10560010@unknown@formal@none@1@S@Morphology (linguistics)@@@@1@2@@danf@17-8-2009
10560020@unknown@formal@none@1@S@'''Morphology''' is the field of [[linguistics]] that studies the internal structure of words.@@@@1@13@@danf@17-8-2009
10560030@unknown@formal@none@1@S@(Words as units in the lexicon are the subject matter of [[lexicology]].)@@@@1@12@@danf@17-8-2009
10560040@unknown@formal@none@1@S@While words are generally accepted as being (with [[clitic]]s) the smallest units of [[syntax]], it is clear that in most (if not all) languages, words can be related to other words by rules.@@@@1@33@@danf@17-8-2009
10560050@unknown@formal@none@1@S@For example, [[English language|English]] speakers recognize that the words ''dog'', ''dogs'', and ''dog-catcher'' are closely related.@@@@1@16@@danf@17-8-2009
10560060@unknown@formal@none@1@S@English speakers recognize these relations from their tacit knowledge of the rules of word-formation in English.@@@@1@16@@danf@17-8-2009
10560070@unknown@formal@none@1@S@They intuit that ''dog'' is to ''dogs'' as ''cat'' is to ''cats''; similarly, ''dog'' is to ''dog-catcher'' as ''dish'' is to ''dishwasher''.@@@@1@22@@danf@17-8-2009
10560080@unknown@formal@none@1@S@The rules understood by the speaker reflect specific patterns (or regularities) in the way words are formed from smaller units and how those smaller units interact in speech.@@@@1@28@@danf@17-8-2009
10560090@unknown@formal@none@1@S@In this way, morphology is the branch of linguistics that studies patterns of word-formation within and across languages, and attempts to formulate rules that model the knowledge of the speakers of those languages.@@@@1@33@@danf@17-8-2009
10560100@unknown@formal@none@1@S@==History ==@@@@1@2@@danf@17-8-2009
10560110@unknown@formal@none@1@S@The history of morphological analysis dates back to the [[ancient India]]n linguist , who formulated the 3,959 rules of [[Sanskrit]] morphology in the text by using a Constituency Grammar.@@@@1@29@@danf@17-8-2009
10560120@unknown@formal@none@1@S@The Graeco-Roman grammatical tradition also engaged in morphological analysis.@@@@1@9@@danf@17-8-2009
10560130@unknown@formal@none@1@S@The term ''morphology'' was coined by [[August Schleicher]] in [[1859]]@@@@1@10@@danf@17-8-2009
10560140@unknown@formal@none@1@S@== Fundamental concepts ==@@@@1@4@@danf@17-8-2009
10560150@unknown@formal@none@1@S@=== Lexemes and word forms ===@@@@1@6@@danf@17-8-2009
10560160@unknown@formal@none@1@S@The distinction between these two senses of "word" is arguably the most important one in morphology.@@@@1@16@@danf@17-8-2009
10560170@unknown@formal@none@1@S@The first sense of "word," the one in which ''dog'' and ''dogs'' are "the same word," is called '''[[lexeme]]'''.@@@@1@19@@danf@17-8-2009
10560180@unknown@formal@none@1@S@The second sense is called '''word-form'''.@@@@1@6@@danf@17-8-2009
10560190@unknown@formal@none@1@S@We thus say that ''dog'' and ''dogs'' are different forms of the same lexeme.@@@@1@14@@danf@17-8-2009
10560200@unknown@formal@none@1@S@''Dog'' and ''dog-catcher'', on the other hand, are different lexemes; for example, they refer to two different kinds of entities.@@@@1@20@@danf@17-8-2009
10560210@unknown@formal@none@1@S@The form of a word that is chosen conventionally to represent the canonical form of a word is called a [[lemma (linguistics)|lemma]], or '''citation form'''.@@@@1@25@@danf@17-8-2009
10560220@unknown@formal@none@1@S@==== Prosodic word vs. morphological word ====@@@@1@7@@danf@17-8-2009
10560230@unknown@formal@none@1@S@Here are examples from other languages of the failure of a single phonological word to coincide with a single morphological word-form.@@@@1@21@@danf@17-8-2009
10560240@unknown@formal@none@1@S@In Latin, one way to express the concept of 'NOUN-PHRASE1 and NOUN-PHRASE2' (as in "apples and oranges") is to suffix '-que' to the second noun phrase: "apples oranges-and", as it were.@@@@1@33@@danf@17-8-2009
10560250@unknown@formal@none@1@S@An extreme level of this theoretical quandary posed by some phonological words is provided by the Kwak'wala language.@@@@1@18@@danf@17-8-2009
10560260@unknown@formal@none@1@S@In Kwak'wala, as in a great many other languages, meaning relations between nouns, including possession and "semantic case", are formulated by affixes instead of by independent "words".@@@@1@27@@danf@17-8-2009
10560270@unknown@formal@none@1@S@The three word English phrase, "with his club", where 'with' identifies its dependent noun phrase as an instrument and 'his' denotes a possession relation, would consist of two words or even just one word in many languages.@@@@1@37@@danf@17-8-2009
10560280@unknown@formal@none@1@S@But affixation for semantic relations in Kwak'wala differs dramatically (from the viewpoint of those whose language is not Kwak'wala) from such affixation in other languages for this reason: the affixes phonologically attach not to the lexeme they pertain to semantically, but to the ''preceding'' lexeme.@@@@1@45@@danf@17-8-2009
10560290@unknown@formal@none@1@S@Consider the following example (in Kwakw'ala, sentences begin with what corresponds to an English verb):@@@@1@15@@danf@17-8-2009
10560300@unknown@formal@none@1@S@kwixʔid-i-da bəgwanəmai-χ-a q'asa-s-isi t'alwagwayu@@@@1@4@@danf@17-8-2009
10560310@unknown@formal@none@1@S@Morpheme by morpheme translation:@@@@1@4@@danf@17-8-2009
10560320@unknown@formal@none@1@S@kwixʔid-i-da = clubbed-PIVOT-DETERMINER@@@@1@4@@danf@17-8-2009
10560330@unknown@formal@none@1@S@bəgwanəma-χ-a = man-ACCUSATIVE-DETERMINER@@@@1@4@@danf@17-8-2009
10560340@unknown@formal@none@1@S@q'asa-s-is = otter-INSTRUMENTAL-3.PERSON.SINGULAR-POSSESSIVE@@@@1@4@@danf@17-8-2009
10560350@unknown@formal@none@1@S@t'alwagwayu = club.@@@@1@3@@danf@17-8-2009
10560360@unknown@formal@none@1@S@"the man clubbed the otter with his club"@@@@1@8@@danf@17-8-2009
10560370@unknown@formal@none@1@S@(Notation notes:@@@@1@2@@danf@17-8-2009
10560380@unknown@formal@none@1@S@1. accusative case marks an entity that something is done to.@@@@1@11@@danf@17-8-2009
10560390@unknown@formal@none@1@S@2. determiners are words such as "the", "this", "that".@@@@1@9@@danf@17-8-2009
10560400@unknown@formal@none@1@S@3. the concept of "pivot" is a theoretical construct that is not relevant to this discussion.)@@@@1@16@@danf@17-8-2009
10560410@unknown@formal@none@1@S@That is, to the speaker of Kwak'wala, the sentence does not contain the "words" 'him-the-otter' or 'with-his-club' Instead, the markers -''i-da'' (PIVOT-'the'), referring to ''man'', attaches not to ''bəgwanəma'' ('man'), but instead to the "verb"; the markers -''χ-a'' (ACCUSATIVE-'the'), referring to ''otter'', attach to ''bəgwanəma'' instead of to ''q'asa'' ('otter'), etc.@@@@1@53@@danf@17-8-2009
10560420@unknown@formal@none@1@S@To summarize differently: a speaker of Kwak'wala does ''not'' perceive the sentence to consist of these phonological words:@@@@1@18@@danf@17-8-2009
10560430@unknown@formal@none@1@S@kwixʔid i-da-bəgwanəma χ-a-q'asa s-isi-t'alwagwayu@@@@1@4@@danf@17-8-2009
10560440@unknown@formal@none@1@S@"clubbed PIVOT-the-mani hit-the-otter with-hisi-club@@@@1@4@@danf@17-8-2009
10560450@unknown@formal@none@1@S@A central publication on this topic is the recent volume edited by Dixon and Aikhenvald (2007), examining the mismatch between prosodic-phonological and grammatical definitions of "word" in various Amazonian, Australian Aboriginal, Caucasian, Eskimo, Indo-European, Native North American, and West African languages, and in sign languages.@@@@1@45@@danf@17-8-2009
10560460@unknown@formal@none@1@S@Apparently, a wide variety of languages make use of the hybrid linguistic unit clitic, possessing the grammatical features of independent words but the prosodic-phonological lack of freedom of bound morphemes.@@@@1@30@@danf@17-8-2009
10560470@unknown@formal@none@1@S@The intermediate status of clitics poses a considerable challenge to linguistic theory.@@@@1@12@@danf@17-8-2009
10560480@unknown@formal@none@1@S@=== Inflection vs. word-formation ===@@@@1@5@@danf@17-8-2009
10560490@unknown@formal@none@1@S@Given the notion of a lexeme, it is possible to distinguish two kinds of morphological rules.@@@@1@16@@danf@17-8-2009
10560500@unknown@formal@none@1@S@Some morphological rules relate to different forms of the same lexeme; while other rules relate to different lexemes.@@@@1@18@@danf@17-8-2009
10560510@unknown@formal@none@1@S@Rules of the first kind are called '''[[Inflection|inflectional rules]]''', while those of the second kind are called '''[[word formation|word-formation]]'''.@@@@1@19@@danf@17-8-2009
10560520@unknown@formal@none@1@S@The English plural, as illustrated by ''dog'' and ''dogs'', is an inflectional rule; compounds like ''dog-catcher'' or ''dishwasher'' provide an example of a word-formation rule.@@@@1@25@@danf@17-8-2009
10560530@unknown@formal@none@1@S@Informally, word-formation rules form "new words" (that is, new lexemes), while inflection rules yield variant forms of the "same" word (lexeme).@@@@1@21@@danf@17-8-2009
10560540@unknown@formal@none@1@S@There is a further distinction between two kinds of word-formation: [[Derivation (linguistics)|derivation]] and [[Compound (linguistics)|compounding]].@@@@1@15@@danf@17-8-2009
10560550@unknown@formal@none@1@S@Compounding is a process of word-formation that involves combining complete word-forms into a single '''compound''' form; ''dog-catcher'' is therefore a compound, because both ''dog'' and ''catcher'' are complete word-forms in their own right before the compounding process has been applied, and are subsequently treated as one form.@@@@1@47@@danf@17-8-2009
10560560@unknown@formal@none@1@S@Derivation involves [[affix]]ing [[bound morpheme|bound]] (non-independent) forms to existing lexemes, whereby the addition of the affix '''derives''' a new lexeme.@@@@1@20@@danf@17-8-2009
10560570@unknown@formal@none@1@S@One example of derivation is clear in this case: the word ''independent'' is derived from the word ''dependent'' by prefixing it with the derivational prefix ''in-'', while ''dependent'' itself is derived from the verb ''depend''.@@@@1@35@@danf@17-8-2009
10560580@unknown@formal@none@1@S@The distinction between inflection and word-formation is not at all clear-cut.@@@@1@11@@danf@17-8-2009
10560590@unknown@formal@none@1@S@There are many examples where linguists fail to agree whether a given rule is inflection or word-formation.@@@@1@17@@danf@17-8-2009
10560600@unknown@formal@none@1@S@The next section will attempt to clarify this distinction.@@@@1@9@@danf@17-8-2009
10560610@unknown@formal@none@1@S@=== Paradigms and morphosyntax ===@@@@1@5@@danf@17-8-2009
10560620@unknown@formal@none@1@S@A '''paradigm''' is the complete set of related word-forms associated with a given lexeme.@@@@1@14@@danf@17-8-2009
10560630@unknown@formal@none@1@S@The familiar examples of paradigms are the [[Grammatical conjugation|conjugations]] of verbs, and the [[declension]]s of nouns.@@@@1@16@@danf@17-8-2009
10560640@unknown@formal@none@1@S@Accordingly, the word-forms of a lexeme may be arranged conveniently into tables, by classifying them according to shared inflectional categories such as [[grammatical tense|tense]], [[grammatical aspect|aspect]], [[grammatical mood|mood]], [[grammatical number|number]], [[grammatical gender|gender]] or [[grammatical case|case]].@@@@1@35@@danf@17-8-2009
10560650@unknown@formal@none@1@S@For example, the personal pronouns in English can be organized into tables, using the categories of person (1st., 2nd., 3rd.), number (singular vs. plural), gender (masculine, feminine, neuter), and [[grammatical case|case]] (subjective, objective, and possessive).@@@@1@35@@danf@17-8-2009
10560660@unknown@formal@none@1@S@See [[English personal pronouns]] for the details.@@@@1@7@@danf@17-8-2009
10560670@unknown@formal@none@1@S@The inflectional categories used to group word-forms into paradigms cannot be chosen arbitrarily; they must be categories that are relevant to stating the [[syntax|syntactic rules]] of the language.@@@@1@28@@danf@17-8-2009
10560680@unknown@formal@none@1@S@For example, person and number are categories that can be used to define paradigms in English, because English has [[Agreement (linguistics)|grammatical agreement]] rules that require the verb in a sentence to appear in an inflectional form that matches the person and number of the subject.@@@@1@45@@danf@17-8-2009
10560690@unknown@formal@none@1@S@In other words, the syntactic rules of English care about the difference between ''dog'' and ''dogs'', because the choice between these two forms determines which form of the verb is to be used.@@@@1@33@@danf@17-8-2009
10560700@unknown@formal@none@1@S@In contrast, however, no syntactic rule of English cares about the difference between ''dog'' and ''dog-catcher'', or ''dependent'' and ''independent''.@@@@1@20@@danf@17-8-2009
10560710@unknown@formal@none@1@S@The first two are just nouns, and the second two just adjectives, and they generally behave like any other noun or adjective behaves.@@@@1@23@@danf@17-8-2009
10560720@unknown@formal@none@1@S@An important difference between inflection and word-formation is that inflected word-forms of lexemes are organized into paradigms, which are defined by the requirements of syntactic rules, whereas the rules of word-formation are not restricted by any corresponding requirements of syntax.@@@@1@40@@danf@17-8-2009
10560730@unknown@formal@none@1@S@Inflection is therefore said to be relevant to syntax, and word-formation is not.@@@@1@13@@danf@17-8-2009
10560740@unknown@formal@none@1@S@The part of morphology that covers the relationship between [[syntax]] and morphology is called morphosyntax, and it concerns itself with inflection and paradigms, but not with word-formation or compounding.@@@@1@29@@danf@17-8-2009
10560750@unknown@formal@none@1@S@=== Allomorphy ===@@@@1@3@@danf@17-8-2009
10560760@unknown@formal@none@1@S@In the exposition above, morphological rules are described as analogies between word-forms: ''dog'' is to ''dogs'' as ''cat'' is to ''cats'', and as ''dish'' is to ''dishes''.@@@@1@27@@danf@17-8-2009
10560770@unknown@formal@none@1@S@In this case, the analogy applies both to the form of the words and to their meaning: in each pair, the first word means "one of X", while the second "two or more of X", and the difference is always the plural form ''-s'' affixed to the second word, signaling the key distinction between singular and plural entities.@@@@1@58@@danf@17-8-2009
10560780@unknown@formal@none@1@S@One of the largest sources of complexity in morphology is that this one-to-one correspondence between meaning and form scarcely applies to every case in the language.@@@@1@26@@danf@17-8-2009
10560790@unknown@formal@none@1@S@In English, we have word form pairs like ''ox/oxen'', ''goose/geese'', and ''sheep/sheep'', where the difference between the singular and the plural is signaled in a way that departs from the regular pattern, or is not signaled at all.@@@@1@38@@danf@17-8-2009
10560800@unknown@formal@none@1@S@Even cases considered "regular", with the final ''-s'', are not so simple; the ''-s'' in ''dogs'' is not pronounced the same way as the ''-s'' in ''cats'', and in a plural like ''dishes'', an "extra" vowel appears before the ''-s''.@@@@1@40@@danf@17-8-2009
10560810@unknown@formal@none@1@S@These cases, where the same distinction is effected by alternative forms of a "word", are called '''[[allomorph]]y'''.@@@@1@17@@danf@17-8-2009
10560820@unknown@formal@none@1@S@Phonological rules constrain which sounds can appear next to each other in a language, and morphological rules, when applied blindly, would often violate phonological rules, by resulting in sound sequences that are prohibited in the language in question.@@@@1@38@@danf@17-8-2009
10560830@unknown@formal@none@1@S@For example, to form the plural of ''dish'' by simply appending an ''-s'' to the end of the word would result in the form *{{IPA|[dɪʃs]}}, which is not permitted by the [[phonotactics]] of English.@@@@1@34@@danf@17-8-2009
10560840@unknown@formal@none@1@S@In order to "rescue" the word, a vowel sound is inserted between the root and the plural marker, and {{IPA|[dɪʃəz]}} results.@@@@1@21@@danf@17-8-2009
10560850@unknown@formal@none@1@S@Similar rules apply to the pronunciation of the ''-s'' in ''dogs'' and ''cats'': it depends on the quality (voiced vs. unvoiced) of the final preceding [[phoneme]].@@@@1@26@@danf@17-8-2009
10560860@unknown@formal@none@1@S@=== Lexical morphology ===@@@@1@4@@danf@17-8-2009
10560870@unknown@formal@none@1@S@[[Lexical morphology]] is the branch of morphology that deals with the [[lexicon]], which, morphologically conceived, is the collection of [[lexeme]]s in a language.@@@@1@23@@danf@17-8-2009
10560880@unknown@formal@none@1@S@As such, it concerns itself primarily with word-formation: derivation and compounding.@@@@1@11@@danf@17-8-2009
10560890@unknown@formal@none@1@S@== Models of morphology ==@@@@1@5@@danf@17-8-2009
10560900@unknown@formal@none@1@S@There are three principal approaches to morphology, which each try to capture the distinctions above in different ways.@@@@1@18@@danf@17-8-2009
10560910@unknown@formal@none@1@S@These are,@@@@1@2@@danf@17-8-2009
10560920@unknown@formal@none@1@S@* [[Morpheme-based morphology]], which makes use of an [[Item-and-Arrangment (Morphology)|Item-and-Arrangement]] approach.@@@@1@11@@danf@17-8-2009
10560930@unknown@formal@none@1@S@* [[Lexeme-based morphology]], which normally makes use of an [[Item-and-Process (Morphology)|Item-and-Process]] approach.@@@@1@12@@danf@17-8-2009
10560940@unknown@formal@none@1@S@* [[Word-based morphology]], which normally makes use of a [[Word-and-paradigm morphology|Word-and-Paradigm]] approach.@@@@1@12@@danf@17-8-2009
10560950@unknown@formal@none@1@S@Note that while the associations indicated between the concepts in each item in that list is very strong, it is not absolute.@@@@1@22@@danf@17-8-2009
10560960@unknown@formal@none@1@S@=== Morpheme-based morphology ===@@@@1@4@@danf@17-8-2009
10560970@unknown@formal@none@1@S@In [[morpheme-based morphology]], word-forms are analyzed as arrangements of [[morpheme]]s.@@@@1@10@@danf@17-8-2009
10560980@unknown@formal@none@1@S@A '''morpheme''' is defined as the minimal meaningful unit of a language.@@@@1@12@@danf@17-8-2009
10560990@unknown@formal@none@1@S@In a word like ''independently'', we say that the morphemes are ''in-'', ''depend'', ''-ent'', and ''ly''; ''depend'' is the [[root (linguistics)|root]] and the other morphemes are, in this case, derivational affixes.@@@@1@31@@danf@17-8-2009
10561000@unknown@formal@none@1@S@In a word like ''dogs'', we say that ''dog'' is the root, and that ''-s'' is an inflectional morpheme.@@@@1@19@@danf@17-8-2009
10561010@unknown@formal@none@1@S@This way of analyzing word-forms as if they were made of morphemes put after each other like beads on a string, is called [[Item-and-Arrangment (Morphology)|Item-and-Arrangement]].@@@@1@25@@danf@17-8-2009
10561020@unknown@formal@none@1@S@The morpheme-based approach is the first one that beginners to morphology usually think of, and which laymen tend to find the most obvious.@@@@1@23@@danf@17-8-2009
10561030@unknown@formal@none@1@S@This is so to such an extent that very often beginners think that morphemes are an inevitable, fundamental notion of morphology, and many five-minute explanations of morphology are, in fact, five-minute explanations of morpheme-based morphology.@@@@1@35@@danf@17-8-2009
10561040@unknown@formal@none@1@S@This is, however, not so.@@@@1@5@@danf@17-8-2009
10561050@unknown@formal@none@1@S@The fundamental idea of morphology is that the words of a language are related to each other by different kinds of rules.@@@@1@22@@danf@17-8-2009
10561060@unknown@formal@none@1@S@Analyzing words as sequences of morphemes is a way of describing these relations, but is not the only way.@@@@1@19@@danf@17-8-2009
10561070@unknown@formal@none@1@S@In actual academic linguistics, morpheme-based morphology certainly has many adherents, but is by no means the dominant approach.@@@@1@18@@danf@17-8-2009
10561080@unknown@formal@none@1@S@=== Lexeme-based morphology ===@@@@1@4@@danf@17-8-2009
10561090@unknown@formal@none@1@S@[[Lexeme-based morphology]] is (usually) an [[Item-and-Process (Morphology)|Item-and-Process]] approach.@@@@1@8@@danf@17-8-2009
10561100@unknown@formal@none@1@S@Instead of analyzing a word-form as a set of morphemes arranged in sequence, a word-form is said to be the result of applying rules that ''alter'' a word-form or stem in order to produce a new one.@@@@1@37@@danf@17-8-2009
10561110@unknown@formal@none@1@S@An inflectional rule takes a stem, changes it as is required by the rule, and outputs a word-form; a derivational rule takes a stem, changes it as per its own requirements, and outputs a derived stem; a compounding rule takes word-forms, and similarly outputs a compound stem.@@@@1@47@@danf@17-8-2009
10561120@unknown@formal@none@1@S@=== Word-based morphology ===@@@@1@4@@danf@17-8-2009
10561130@unknown@formal@none@1@S@[[Word-based morphology]] is a (usually) [[Word-and-paradigm morphology|Word-and-paradigm]] approach.@@@@1@8@@danf@17-8-2009
10561140@unknown@formal@none@1@S@This theory takes paradigms as a central notion.@@@@1@8@@danf@17-8-2009
10561150@unknown@formal@none@1@S@Instead of stating rules to combine morphemes into word-forms, or to generate word-forms from stems, word-based morphology states generalizations that hold between the forms of inflectional paradigms.@@@@1@27@@danf@17-8-2009
10561160@unknown@formal@none@1@S@The major point behind this approach is that many such generalizations are hard to state with either of the other approaches.@@@@1@21@@danf@17-8-2009
10561170@unknown@formal@none@1@S@The examples are usually drawn from [[fusional language]]s, where a given "piece" of a word, which a morpheme-based theory would call an inflectional morpheme, corresponds to a combination of grammatical categories, for example, "third person plural."@@@@1@36@@danf@17-8-2009
10561180@unknown@formal@none@1@S@Morpheme-based theories usually have no problems with this situation, since one just says that a given morpheme has two categories.@@@@1@20@@danf@17-8-2009
10561190@unknown@formal@none@1@S@Item-and-Process theories, on the other hand, often break down in cases like these, because they all too often assume that there will be two separate rules here, one for third person, and the other for plural, but the distinction between them turns out to be artificial.@@@@1@46@@danf@17-8-2009
10561200@unknown@formal@none@1@S@Word-and-Paradigm approaches treat these as whole words that are related to each other by [[analogy|analogical]] rules.@@@@1@16@@danf@17-8-2009
10561210@unknown@formal@none@1@S@Words can be categorized based on the pattern they fit into.@@@@1@11@@danf@17-8-2009
10561220@unknown@formal@none@1@S@This applies both to existing words and to new ones.@@@@1@10@@danf@17-8-2009
10561230@unknown@formal@none@1@S@Application of a pattern different than the one that has been used historically can give rise to a new word, such as ''older'' replacing ''elder'' (where ''older'' follows the normal pattern of [[adjective|adjectival]] [[superlative]]s) and ''cows'' replacing ''kine'' (where ''cows'' fits the regular pattern of plural formation).@@@@1@47@@danf@17-8-2009
10561240@unknown@formal@none@1@S@While a Word-and-Paradigm approach can explain this easily, other approaches have difficulty with phenomena such as this.@@@@1@17@@danf@17-8-2009
10561250@unknown@formal@none@1@S@== Morphological typology ==@@@@1@4@@danf@17-8-2009
10561260@unknown@formal@none@1@S@In the 19th century, philologists devised a now classic classification of languages according to their morphology.@@@@1@16@@danf@17-8-2009
10561270@unknown@formal@none@1@S@According to this typology, some languages are [[isolating language|isolating]], and have little to no morphology; others are [[agglutinating language|agglutinative]], and their words tend to have lots of easily-separable morphemes; while others yet are inflectional or [[fusional language|fusional]], because their inflectional morphemes are said to be "fused" together.@@@@1@47@@danf@17-8-2009
10561280@unknown@formal@none@1@S@This leads to one bound morpheme conveying multiple pieces of information.@@@@1@11@@danf@17-8-2009
10561290@unknown@formal@none@1@S@The classic example of an isolating language is [[Chinese language|Chinese]]; the classic example of an agglutinative language is [[Turkish language|Turkish]]; both [[Latin language|Latin]] and [[Greek language|Greek]] are classic examples of fusional languages.@@@@1@32@@danf@17-8-2009
10561300@unknown@formal@none@1@S@Considering the variability of the world's languages, it becomes clear that this classification is not at all clear-cut, and many languages do not neatly fit any one of these types, and some fit in more than one.@@@@1@37@@danf@17-8-2009
10561310@unknown@formal@none@1@S@A continuum of complex morphology of language may be adapted when considering languages.@@@@1@13@@danf@17-8-2009
10561320@unknown@formal@none@1@S@The three models of morphology stem from attempts to analyze languages that more or less match different categories in this typology.@@@@1@21@@danf@17-8-2009
10561330@unknown@formal@none@1@S@The Item-and-Arrangement approach fits very naturally with agglutinative languages; while the Item-and-Process and Word-and-Paradigm approaches usually address fusional languages.@@@@1@19@@danf@17-8-2009
10561340@unknown@formal@none@1@S@The reader should also note that the classical typology also mostly applies to inflectional morphology.@@@@1@15@@danf@17-8-2009
10561350@unknown@formal@none@1@S@There is very little fusion going on with word-formation.@@@@1@9@@danf@17-8-2009
10561360@unknown@formal@none@1@S@Languages may be classified as synthetic or analytic in their word formation, depending on the preferred way of expressing notions that are not inflectional: either by using word-formation (synthetic), or by using syntactic phrases (analytic).@@@@1@35@@danf@17-8-2009
10570010@unknown@formal@none@1@S@Named entity recognition@@@@1@3@@danf@17-8-2009
10570020@unknown@formal@none@1@S@'''Named entity recognition''' (NER) (also known as '''entity identification (EI)''' and '''entity extraction''') is a subtask of [[information extraction]] that seeks to locate and classify atomic elements in text into predefined categories such as the names of persons, organizations, locations, expressions of times, quantities, monetary values, percentages, etc.@@@@1@48@@danf@17-8-2009
10570030@unknown@formal@none@1@S@For example, a NER system producing [[Message Understanding Conference|MUC]]-style output might [[Metadata|tag]] the sentence,@@@@1@14@@danf@17-8-2009
10570040@unknown@formal@none@1@S@:''Jim bought 300 shares of Acme Corp. in 2006.''@@@@1@9@@danf@17-8-2009
10570050@unknown@formal@none@1@S@:''''''''Jim'''''' bought ''''''300'''''' shares of ''''''Acme Corp.'''''' in ''''''2006''''''''.@@@@1@13@@danf@17-8-2009
10570060@unknown@formal@none@1@S@NER systems have been created that use linguistic [[formal grammar|grammar]]-based techniques as well as [[statistical model]]s.@@@@1@16@@danf@17-8-2009
10570070@unknown@formal@none@1@S@Hand-crafted grammar-based systems typically obtain better results, but at the cost of months of work by experienced [[Linguistics|linguists]].@@@@1@18@@danf@17-8-2009
10570080@unknown@formal@none@1@S@Statistical NER systems typically require a large amount of manually [[annotation|annotated]] training data.@@@@1@13@@danf@17-8-2009
10570090@unknown@formal@none@1@S@Since about 1998, there has been a great deal of interest in entity identification in the [[molecular biology]], [[bioinformatics]], and medical [[natural language processing]] communities.@@@@1@25@@danf@17-8-2009
10570100@unknown@formal@none@1@S@The most common entity of interest in that domain has been names of genes and gene products.@@@@1@17@@danf@17-8-2009
10570110@unknown@formal@none@1@S@==Named entity types==@@@@1@3@@danf@17-8-2009
10570120@unknown@formal@none@1@S@In the expression ''named entity'', the word ''named'' restricts the task to those entities for which one or many [[rigid designator]]s, as defined by [[Saul Kripke|Kripke]], stands for the referent.@@@@1@30@@danf@17-8-2009
10570130@unknown@formal@none@1@S@For instance, the ''automotive company created by Henry Ford in 1903'' is referred to as ''Ford'' or ''Ford Motor Company''.@@@@1@20@@danf@17-8-2009
10570140@unknown@formal@none@1@S@Rigid designators include proper names as well as certain natural kind terms like biological species and substances.@@@@1@17@@danf@17-8-2009
10570150@unknown@formal@none@1@S@There is a general agreement to include [[temporal expressions]] and some numerical expressions such as money and measures in named entities.@@@@1@21@@danf@17-8-2009
10570160@unknown@formal@none@1@S@While some instances of these types are good examples of rigid designators (e.g., the year 2001) there are also many invalid ones (e.g., I take my vacations in “June”).@@@@1@29@@danf@17-8-2009
10570170@unknown@formal@none@1@S@In the first case, the year ''2001'' refers to the ''2001st year of the Gregorian calendar''.@@@@1@16@@danf@17-8-2009
10570180@unknown@formal@none@1@S@In the second case, the month ''June'' may refer to the month of an undefined year (''past June'', ''next June'', ''June 2020'', etc.).@@@@1@23@@danf@17-8-2009
10570190@unknown@formal@none@1@S@It is arguable that the named entity definition is loosened in such cases for practical reasons.@@@@1@16@@danf@17-8-2009
10570200@unknown@formal@none@1@S@At least two [[Hierarchy|hierarchies]] of named entity types have been proposed in the literature.@@@@1@14@@danf@17-8-2009
10570210@unknown@formal@none@1@S@[[BBN Technologies|BBN]] categories [http://www.ldc.upenn.edu/Catalog/docs/LDC2005T33/BBN-Types-Subtypes.html], proposed in 2002, is used for [[Question Answering]] and consists of 29 types and 64 subtypes.@@@@1@20@@danf@17-8-2009
10570220@unknown@formal@none@1@S@Sekine's extended hierarchy [http://nlp.cs.nyu.edu/ene/], proposed in 2002, is made of 200 subtypes.@@@@1@12@@danf@17-8-2009
10570230@unknown@formal@none@1@S@==Evaluation==@@@@1@1@@danf@17-8-2009
10570240@unknown@formal@none@1@S@Benchmarking and evaluations have been performed in the ''[[Message Understanding Conference]]s'' (MUC) organized by [[DARPA]], ''International Conference on Language Resources and Evaluation (LREC)'', ''Computational Natural Language Learning ([[CoNLL]])'' workshops, ''Automatic Content Extraction'' (ACE) organized by [[NIST]], the ''[[Multilingual Entity Task Conference]]'' (MET), ''Information Retrieval and Extraction Exercise'' (IREX) and in ''HAREM'' (Portuguese language only).@@@@1@54@@danf@17-8-2009
10570250@unknown@formal@none@1@S@[http://aclweb.org/aclwiki/index.php?title=Named_Entity_Recognition_%28State_of_the_art%29 State-of-the-art systems] produce near-human performance.@@@@1@6@@danf@17-8-2009
10570260@unknown@formal@none@1@S@For instance, the best system entering [http://www.itl.nist.gov/iad/894.02/related_projects/muc/proceedings/muc_7_toc.html MUC-7] scored 93.39% of [[Information_retrieval#F-measure|f-measure]] while human annotators scored 97.60% and 96.95%.@@@@1@19@@danf@17-8-2009
10580010@unknown@formal@none@1@S@Natural language@@@@1@2@@danf@17-8-2009
10580020@unknown@formal@none@1@S@In the [[philosophy of language]], a '''natural language''' (or '''ordinary language''') is a [[language]] that is spoken, [[writing|written]], or [[Sign language|signed]] by [[animal]]s for general-purpose communication, as distinguished from [[formal language]]s (such as [[Programming language|computer-programming languages]] or the "languages" used in the study of formal [[logic]], especially [[mathematical logic]]) and from [[constructed language]]s.@@@@1@53@@danf@17-8-2009
10580030@unknown@formal@none@1@S@== Defining natural language ==@@@@1@5@@danf@17-8-2009
10580040@unknown@formal@none@1@S@Though the exact definition is debatable, natural language is often contrasted with artificial or [[constructed languages]] such as [[Esperanto]], [[Latino Sexione]], and [[Occidental language|Occidental]].@@@@1@24@@danf@17-8-2009
10580050@unknown@formal@none@1@S@Linguists have an incomplete understanding of all aspects of the rules underlying natural languages, and these rules are therefore objects of study.@@@@1@22@@danf@17-8-2009
10580060@unknown@formal@none@1@S@The understanding of natural languages reveals much about not only how language works (in terms of [[syntax]], [[semantics]], [[phonetics]], [[phonology]], etc), but also about how the human [[mind]] and the human [[brain]] process language.@@@@1@34@@danf@17-8-2009
10580070@unknown@formal@none@1@S@In linguistic terms, 'natural language' only applies to a language that has evolved naturally, and the study of natural language primarily involves native (first language) speakers.@@@@1@26@@danf@17-8-2009
10580080@unknown@formal@none@1@S@The theory of [[universal grammar]] proposes that all natural languages have certain underlying rules which constrain the structure of the specific grammar for any given language.@@@@1@26@@danf@17-8-2009
10580090@unknown@formal@none@1@S@While [[grammarians]], writers of dictionaries, and language policy-makers all have a certain influence on the evolution of language, their ability to influence what people think they 'ought' to say is distinct from what people actually say.@@@@1@36@@danf@17-8-2009
10580100@unknown@formal@none@1@S@Natural language applies to the latter, and is thus a 'descriptive' rather than a 'prescriptive' term.@@@@1@16@@danf@17-8-2009
10580110@unknown@formal@none@1@S@Thus non-standard language varieties (such as [[African American Vernacular English]]) are considered to be natural while standard language varieties (such as [[Standard American English]]) which are more 'prescripted' can be considered to be at least somewhat artificial or constructed.@@@@1@39@@danf@17-8-2009
10580120@unknown@formal@none@1@S@== Native language learning ==@@@@1@5@@danf@17-8-2009
10580130@unknown@formal@none@1@S@The [[learning]] of one's own [[native language]], typically that of one's [[parent]]s, normally occurs spontaneously in early human [[childhood]] and is [[Biology|biologically]] driven.@@@@1@23@@danf@17-8-2009
10580140@unknown@formal@none@1@S@A crucial role of this process is performed by the [[Nervous system|neural]] activity of a portion of the human [[brain]] known as [[Broca's area]].@@@@1@24@@danf@17-8-2009
10580150@unknown@formal@none@1@S@There are approximately 7,000 current human languages, and many, if not most seem to share certain properties, leading to the belief in the existence of [[Universal Grammar]], as shown by [[generative grammar]] studies pioneered by the work of [[Noam Chomsky]].@@@@1@40@@danf@17-8-2009
10580160@unknown@formal@none@1@S@Recently, it has been demonstrated that a dedicated network in the human brain (crucially involving [[Broca's area]], a portion of the left inferior frontal gyrus), is selectively activated by complex verbal structures (but not simple ones) of those languages that meet the Universal Grammar requirements.@@@@1@45@@danf@17-8-2009
10580170@unknown@formal@none@1@S@== Origins of natural language ==@@@@1@6@@danf@17-8-2009
10580180@unknown@formal@none@1@S@There is disagreement among anthropologists on when language was first used by humans (or their ancestors).@@@@1@16@@danf@17-8-2009
10580190@unknown@formal@none@1@S@Estimates range from about two million (2,000,000) years ago, during the time of ''[[Homo habilis]]'', to as recently as forty thousand (40,000) years ago, during the time of [[Cro-Magnon]] man.@@@@1@30@@danf@17-8-2009
10580200@unknown@formal@none@1@S@However recent evidence suggests modern human language was invented or evolved in Africa prior to the dispersal of humans from Africa around 50,000 years ago.@@@@1@25@@danf@17-8-2009
10580210@unknown@formal@none@1@S@Since all people including the most isolated indigenous groups such as the [[Andamanese]] or the [[Tasmanian aboriginals]] possess language, then it must have been present in the ancestral populations in Africa before the human population split into various groups to colonize the rest of the world.@@@@1@46@@danf@17-8-2009
10580220@unknown@formal@none@1@S@Some claim that all nautural languages came out of one single language, known as [[Adamic]].@@@@1@15@@danf@17-8-2009
10580230@unknown@formal@none@1@S@== Linguistic diversity ==@@@@1@4@@danf@17-8-2009
10580240@unknown@formal@none@1@S@As of early 2007, there are 6,912 known living human languages.@@@@1@11@@danf@17-8-2009
10580250@unknown@formal@none@1@S@A "living language" is simply one which is in wide use by a specific group of living people.@@@@1@18@@danf@17-8-2009
10580260@unknown@formal@none@1@S@The exact number of known living languages will vary from 5,000 to 10,000, depending generally on the precision of one's definition of "language", and in particular on how one classifies [[dialects]].@@@@1@31@@danf@17-8-2009
10580270@unknown@formal@none@1@S@There are also many dead or [[extinct language]]s.@@@@1@8@@danf@17-8-2009
10580280@unknown@formal@none@1@S@There is no [[dialect#.22Dialect.22 or .22language.22|clear distinction]] between a language and a [[dialect]], notwithstanding linguist [[Max Weinreich]]'s famous [[aphorism]] that "[[a language is a dialect with an army and navy]]."@@@@1@30@@danf@17-8-2009
10580290@unknown@formal@none@1@S@In other words, the distinction may hinge on political considerations as much as on cultural differences, distinctive [[writing system]]s, or degree of [[mutual intelligibility]].@@@@1@24@@danf@17-8-2009
10580300@unknown@formal@none@1@S@It is probably impossible to accurately enumerate the living languages because our worldwide knowledge is incomplete, and it is a "moving target", as explained in greater detail by the [[Ethnologue]]'s Introduction, p. 7 - 8.@@@@1@35@@danf@17-8-2009
10580310@unknown@formal@none@1@S@With the 15th edition, the 103 newly added languages are not new but reclassified due to refinements in the definition of language.@@@@1@22@@danf@17-8-2009
10580320@unknown@formal@none@1@S@Although widely considered an [[encyclopedia]], the [[Ethnologue]] actually presents itself as an incomplete catalog, including only named languages that its editors are able to document.@@@@1@25@@danf@17-8-2009
10580330@unknown@formal@none@1@S@With each edition, the number of catalogued languages has grown.@@@@1@10@@danf@17-8-2009
10580340@unknown@formal@none@1@S@Beginning with the 14th edition (2000), an attempt was made to include all known living languages.@@@@1@16@@danf@17-8-2009
10580350@unknown@formal@none@1@S@SIL used an internal 3-letter code fashioned after [[airport code]]s to identify languages.@@@@1@13@@danf@17-8-2009
10580360@unknown@formal@none@1@S@This was the precursor to the modern [[ISO 639-3]] standard, to which SIL contributed.@@@@1@14@@danf@17-8-2009
10580370@unknown@formal@none@1@S@The standard allows for over 14,000 languages.@@@@1@7@@danf@17-8-2009
10580380@unknown@formal@none@1@S@In turn, the 15th edition was revised to conform to the pending ISO 639-3 standard.@@@@1@15@@danf@17-8-2009
10580390@unknown@formal@none@1@S@Of the catalogued languages, 497 have been flagged as "nearly extinct" due to trends in their usage.@@@@1@17@@danf@17-8-2009
10580400@unknown@formal@none@1@S@Per the 15th edition, 6,912 living languages are shared by over 5.7 billion speakers. (p. 15)@@@@1@16@@danf@17-8-2009
10580410@unknown@formal@none@1@S@== Taxonomy ==@@@@1@3@@danf@17-8-2009
10580420@unknown@formal@none@1@S@The [[Taxonomic classification|classification]] of natural languages can be performed on the basis of different underlying principles (different closeness notions, respecting different properties and relations between languages); important directions of present classifications are:@@@@1@32@@danf@17-8-2009
10580430@unknown@formal@none@1@S@* paying attention to the historical evolution of languages results in a genetic classification of languages—which is based on genetic relatedness of languages,@@@@1@23@@danf@17-8-2009
10580440@unknown@formal@none@1@S@* paying attention to the internal structure of languages ([[grammar]]) results in a typological classification of languages—which is based on similarity of one or more components of the language's grammar across languages,@@@@1@32@@danf@17-8-2009
10580450@unknown@formal@none@1@S@* and respecting geographical closeness and contacts between language-speaking communities results in areal groupings of languages.@@@@1@16@@danf@17-8-2009
10580460@unknown@formal@none@1@S@The different classifications do not match each other and are not expected to, but the correlation between them is an important point for many [[linguistics|linguistic]] research works.@@@@1@27@@danf@17-8-2009
10580470@unknown@formal@none@1@S@(There is a parallel to the classification of [[species]] in biological [[phylogenetics]] here: consider [[monophyletic]] vs. [[polyphyletic]] groups of species.)@@@@1@20@@danf@17-8-2009
10580480@unknown@formal@none@1@S@The task of genetic classification belongs to the field of [[historical-comparative linguistics]], of typological—to [[linguistic typology]].@@@@1@16@@danf@17-8-2009
10580490@unknown@formal@none@1@S@See also [[Taxonomy]], and [[Taxonomic classification]] for the general idea of classification and taxonomies.@@@@1@14@@danf@17-8-2009
10580500@unknown@formal@none@1@S@==== Genetic classification ====@@@@1@4@@danf@17-8-2009
10580510@unknown@formal@none@1@S@The world's languages have been grouped into families of languages that are believed to have common ancestors.@@@@1@17@@danf@17-8-2009
10580520@unknown@formal@none@1@S@Some of the major families are the [[Indo-European languages]], the [[Afro-Asiatic languages]], the [[Austronesian languages]], and the [[Sino-Tibetan languages]].@@@@1@19@@danf@17-8-2009
10580530@unknown@formal@none@1@S@The shared features of languages from one family can be due to shared ancestry.@@@@1@14@@danf@17-8-2009
10580540@unknown@formal@none@1@S@(Compare with [[homology (biology)|homology]] in biology.)@@@@1@6@@danf@17-8-2009
10580550@unknown@formal@none@1@S@==== Typological classification ====@@@@1@4@@danf@17-8-2009
10580560@unknown@formal@none@1@S@An example of a typological classification is the classification of languages on the basis of the basic order of the [[verb]], the [[subject (grammar)|subject]] and the [[object (grammar)|object]] in a [[sentence (linguistics)|sentence]] into several types: [[SVO language|SVO]], [[SOV language|SOV]], [[VSO language|VSO]], and so on, languages.@@@@1@45@@danf@17-8-2009
10580570@unknown@formal@none@1@S@([[English language|English]], for instance, belongs to the [[SVO language]] type.)@@@@1@10@@danf@17-8-2009
10580580@unknown@formal@none@1@S@The shared features of languages of one type (= from one typological class) may have arisen completely independently.@@@@1@18@@danf@17-8-2009
10580590@unknown@formal@none@1@S@(Compare with [[analogy (biology)|analogy]] in biology.)@@@@1@6@@danf@17-8-2009
10580595@unknown@formal@none@1@S@Their cooccurence might be due to the universal laws governing the structure of natural languages—[[language universal]]s.@@@@1@16@@danf@17-8-2009
10580600@unknown@formal@none@1@S@==== Areal classification ====@@@@1@4@@danf@17-8-2009
10580610@unknown@formal@none@1@S@The following language groupings can serve as some linguistically significant examples of areal linguistic units, or ''[[sprachbund]]s'': [[Balkan linguistic union]], or the bigger group of [[European languages]]; [[Caucasian languages]]; [[East Asian languages]].@@@@1@32@@danf@17-8-2009
10580620@unknown@formal@none@1@S@Although the members of each group are not closely [[genetic relatedness of languages|genetically related]], there is a reason for them to share similar features, namely: their speakers have been in contact for a long time within a common community and the languages ''converged'' in the course of the history.@@@@1@49@@danf@17-8-2009
10580630@unknown@formal@none@1@S@These are called "[[areal feature (linguistics)|areal feature]]s".@@@@1@7@@danf@17-8-2009
10580640@unknown@formal@none@1@S@One should be careful about the underlying classification principle for groups of languages which have apparently a geographical name: besides areal linguistic units, the [[taxa]] of the genetic classification ([[language family|language families]]) are often given names which themselves or parts of which refer to geographical areas.@@@@1@46@@danf@17-8-2009
10580650@unknown@formal@none@1@S@== Controlled languages ==@@@@1@4@@danf@17-8-2009
10580660@unknown@formal@none@1@S@Controlled natural languages are subsets of natural languages whose grammars and dictionaries have been restricted in order to reduce or eliminate both ambiguity and complexity.@@@@1@25@@danf@17-8-2009
10580670@unknown@formal@none@1@S@The purpose behind the development and implementation of a controlled natural language typically is to aid non-native speakers of a natural language in understanding it, or to ease computer processing of a natural language.@@@@1@34@@danf@17-8-2009
10580680@unknown@formal@none@1@S@An example of a widely used controlled natural language is [[Simplified English]], which was originally developed for [[aerospace]] industry maintenance manuals.@@@@1@21@@danf@17-8-2009
10580690@unknown@formal@none@1@S@== Constructed languages and international auxiliary languages ==@@@@1@8@@danf@17-8-2009
10580700@unknown@formal@none@1@S@Constructed [[international auxiliary language]]s such as [[Esperanto]] and [[Interlingua]] that have [[native speaker]]s are by some also considered natural languages.@@@@1@20@@danf@17-8-2009
10580710@unknown@formal@none@1@S@However, constructed languages, while they are clearly languages, are not generally considered natural languages.@@@@1@14@@danf@17-8-2009
10580720@unknown@formal@none@1@S@The problem is that other languages have been used to communicate and evolve in a natural way, while Esperanto has been selectively designed by [[L.L. Zamenhof]] from natural languages, not grown from the natural fluctuations in vocabulary and syntax.@@@@1@39@@danf@17-8-2009
10580730@unknown@formal@none@1@S@Nor has Esperanto been naturally "standardized" by children's natural tendency to correct for illogical grammar structures in their parents' language, which can be seen in the development of [[pidgin]] languages into [[creole language]]s (as explained by Steven Pinker in [[The Language Instinct]]).@@@@1@42@@danf@17-8-2009
10580740@unknown@formal@none@1@S@The possible exception to this are true native speakers of such languages.@@@@1@12@@danf@17-8-2009
10580750@unknown@formal@none@1@S@More substantive basis for this designation is that the vocabulary, grammar, and orthography of Interlingua are natural; they have been standardized and presented by a [[International Auxiliary Language Association|linguistic research body]], but they predated it and are not themselves considered a product of human invention.@@@@1@45@@danf@17-8-2009
10580760@unknown@formal@none@1@S@Most experts, however, consider Interlingua to be naturalistic rather than natural.@@@@1@11@@danf@17-8-2009
10580770@unknown@formal@none@1@S@[[Latino Sine Flexione]], a second naturalistic auxiliary language, is also naturalistic in content but is no longer widely spoken.@@@@1@19@@danf@17-8-2009
10580780@unknown@formal@none@1@S@==Natural Language Processing==@@@@1@3@@danf@17-8-2009
10580790@unknown@formal@none@1@S@Natural language processing (NLP) is a subfield of artificial intelligence and computational linguistics.@@@@1@13@@danf@17-8-2009
10580800@unknown@formal@none@1@S@It studies the problems of automated generation and understanding of natural human languages.@@@@1@13@@danf@17-8-2009
10580810@unknown@formal@none@1@S@Natural-language-generation systems convert information from computer databases into normal-sounding human language.@@@@1@11@@danf@17-8-2009
10580820@unknown@formal@none@1@S@Natural-language-understanding systems convert samples of human language into more formal representations that are easier for computer programs to manipulate.@@@@1@19@@danf@17-8-2009
10580830@unknown@formal@none@1@S@== Modalities ==@@@@1@3@@danf@17-8-2009
10580840@unknown@formal@none@1@S@Natural language manifests itself in modalities other than speech.@@@@1@9@@danf@17-8-2009
10580850@unknown@formal@none@1@S@=== Sign languages ===@@@@1@4@@danf@17-8-2009
10580860@unknown@formal@none@1@S@In linguistic terms, sign languages are as rich and complex as any oral language, despite the previously common misconception that they are not "real languages".@@@@1@25@@danf@17-8-2009
10580870@unknown@formal@none@1@S@Professional linguists have studied many sign languages and found them to have every linguistic component required to be classed as true natural languages.@@@@1@23@@danf@17-8-2009
10580880@unknown@formal@none@1@S@Sign languages are not [[pantomime]], much as most spoken language is not [[onomatopoeic]].@@@@1@13@@danf@17-8-2009
10580890@unknown@formal@none@1@S@The signs do tend to exploit iconicity (visual connections with their referents) more than what is common in spoken language, but they are above all conventional and hence generally incomprehensible to non-speakers, just like spoken words and morphemes.@@@@1@38@@danf@17-8-2009
10580900@unknown@formal@none@1@S@They are not a visual rendition of an oral language either.@@@@1@11@@danf@17-8-2009
10580910@unknown@formal@none@1@S@They have complex grammars of their own, and can be used to discuss any topic, from the simple and concrete to the lofty and abstract.@@@@1@25@@danf@17-8-2009
10580920@unknown@formal@none@1@S@=== Written languages ===@@@@1@4@@danf@17-8-2009
10580930@unknown@formal@none@1@S@In a sense, written language should be distinguished from natural language.@@@@1@11@@danf@17-8-2009
10580940@unknown@formal@none@1@S@Until recently in the developed world, it was common for many people to be fluent in [[spoken language|spoken]] or [[sign language|signed languages]] and yet remain illiterate; this is still the case in poor countries today.@@@@1@35@@danf@17-8-2009
10580950@unknown@formal@none@1@S@Furthermore, natural [[language acquisition]] during childhood is largely spontaneous, while [[literacy]] must usually be intentionally acquired.@@@@1@16@@danf@17-8-2009
10590010@unknown@formal@none@1@S@Natural language processing@@@@1@3@@danf@17-8-2009
10590020@unknown@formal@none@1@S@'''Natural language processing''' ('''NLP''') is a subfield of [[artificial intelligence]] and [[computational linguistics]].@@@@1@13@@danf@17-8-2009
10590030@unknown@formal@none@1@S@It studies the problems of automated generation and understanding of [[natural language|natural human languages]].@@@@1@14@@danf@17-8-2009
10590040@unknown@formal@none@1@S@Natural-language-generation systems convert information from computer databases into normal-sounding human language.@@@@1@11@@danf@17-8-2009
10590050@unknown@formal@none@1@S@Natural-language-understanding systems convert samples of human language into more formal representations that are easier for [[computer]] programs to manipulate.@@@@1@19@@danf@17-8-2009
10590060@unknown@formal@none@1@S@==Tasks and limitations==@@@@1@3@@danf@17-8-2009
10590070@unknown@formal@none@1@S@In theory, natural-language processing is a very attractive method of [[human-computer interaction]].@@@@1@12@@danf@17-8-2009
10590080@unknown@formal@none@1@S@Early systems such as [[SHRDLU]], working in restricted "[[blocks world]]s" with restricted vocabularies, worked extremely well, leading researchers to excessive optimism, which was soon lost when the systems were extended to more realistic situations with real-world [[ambiguity]] and [[complexity]].@@@@1@39@@danf@17-8-2009
10590090@unknown@formal@none@1@S@Natural-language understanding is sometimes referred to as an [[AI-complete]] problem, because natural-language recognition seems to require extensive knowledge about the outside world and the ability to manipulate it.@@@@1@28@@danf@17-8-2009
10590100@unknown@formal@none@1@S@The definition of "[[understanding]]" is one of the major problems in natural-language processing.@@@@1@13@@danf@17-8-2009
10590110@unknown@formal@none@1@S@==Concrete problems==@@@@1@2@@danf@17-8-2009
10590120@unknown@formal@none@1@S@Some examples of the problems faced by natural-language-understanding systems:@@@@1@9@@danf@17-8-2009
10590130@unknown@formal@none@1@S@* The sentences ''We gave the monkeys the bananas because they were hungry'' and ''We gave the monkeys the bananas because they were over-ripe'' have the same surface grammatical structure.@@@@1@30@@danf@17-8-2009
10590140@unknown@formal@none@1@S@However, the pronoun ''they'' refers to ''monkeys'' in one sentence and ''bananas'' in the other, and it is impossible to tell which without a knowledge of the properties of monkeys and bananas.@@@@1@32@@danf@17-8-2009
10590150@unknown@formal@none@1@S@* A string of words may be interpreted in different ways.@@@@1@11@@danf@17-8-2009
10590160@unknown@formal@none@1@S@For example, the string ''Time flies like an arrow'' may be interpreted in a variety of ways:@@@@1@17@@danf@17-8-2009
10590170@unknown@formal@none@1@S@**The common [[simile]]: ''[[time]]'' moves quickly just like an arrow does;@@@@1@11@@danf@17-8-2009
10590180@unknown@formal@none@1@S@**measure the speed of flies like you would measure that of an arrow (thus interpreted as an imperative) - i.e. ''(You should) time flies as you would (time) an arrow.'';@@@@1@30@@danf@17-8-2009
10590190@unknown@formal@none@1@S@**measure the speed of flies like an arrow would - i.e. ''Time flies in the same way that an arrow would (time them).'';@@@@1@23@@danf@17-8-2009
10590200@unknown@formal@none@1@S@**measure the speed of flies that are like arrows - i.e. ''Time those flies that are like arrows'';@@@@1@18@@danf@17-8-2009
10590210@unknown@formal@none@1@S@**all of a type of flying insect, "time-flies," collectively enjoys a single arrow (compare ''Fruit flies like a banana'');@@@@1@19@@danf@17-8-2009
10590220@unknown@formal@none@1@S@**each of a type of flying insect, "time-flies," individually enjoys a different arrow (similar comparison applies);@@@@1@16@@danf@17-8-2009
10590230@unknown@formal@none@1@S@**A concrete object, for example the magazine, ''[[Time (magazine)|Time]]'', travels through the air in an arrow-like manner.@@@@1@17@@danf@17-8-2009
10590240@unknown@formal@none@1@S@English is particularly challenging in this regard because it has little [[inflectional morphology]] to distinguish between [[parts of speech]].@@@@1@19@@danf@17-8-2009
10590250@unknown@formal@none@1@S@* English and several other languages don't specify which word an adjective applies to.@@@@1@14@@danf@17-8-2009
10590260@unknown@formal@none@1@S@For example, in the string "pretty little girls' school".@@@@1@9@@danf@17-8-2009
10590270@unknown@formal@none@1@S@** Does the school look little?@@@@1@6@@danf@17-8-2009
10590280@unknown@formal@none@1@S@** Do the girls look little?@@@@1@6@@danf@17-8-2009
10590290@unknown@formal@none@1@S@** Do the girls look pretty?@@@@1@6@@danf@17-8-2009
10590300@unknown@formal@none@1@S@** Does the school look pretty?@@@@1@6@@danf@17-8-2009
10590310@unknown@formal@none@1@S@* We will often imply additional information in spoken language by the way we place stress on words.@@@@1@18@@danf@17-8-2009
10590320@unknown@formal@none@1@S@The sentence "I never said she stole my money" demonstrates the importance stress can play in a sentence, and thus the inherent difficulty a natural language processor can have in parsing it.@@@@1@32@@danf@17-8-2009
10590330@unknown@formal@none@1@S@Depending on which word the speaker places the stress, this sentence could have several distinct meanings:@@@@1@16@@danf@17-8-2009
10590340@unknown@formal@none@1@S@** "'''I''' never said she stole my money" - Someone else said it, but ''I'' didn't.@@@@1@16@@danf@17-8-2009
10590350@unknown@formal@none@1@S@** "I '''never''' said she stole my money" - I simply didn't ever say it.@@@@1@15@@danf@17-8-2009
10590360@unknown@formal@none@1@S@** "I never '''said''' she stole my money" - I might have implied it in some way, but I never explicitly said it.@@@@1@23@@danf@17-8-2009
10590370@unknown@formal@none@1@S@** "I never said '''she''' stole my money" - I said someone took it; I didn't say it was she.@@@@1@20@@danf@17-8-2009
10590380@unknown@formal@none@1@S@** "I never said she '''stole''' my money" - I just said she probably borrowed it.@@@@1@16@@danf@17-8-2009
10590390@unknown@formal@none@1@S@** "I never said she stole '''my''' money" - I said she stole someone else's money.@@@@1@16@@danf@17-8-2009
10590400@unknown@formal@none@1@S@** "I never said she stole my '''money'''" - I said she stole something, but not my money.@@@@1@18@@danf@17-8-2009
10590410@unknown@formal@none@1@S@==Subproblems==@@@@1@1@@danf@17-8-2009
10590420@unknown@formal@none@1@S@; [[Speech segmentation]]:@@@@1@3@@danf@17-8-2009
10590430@unknown@formal@none@1@S@In most spoken languages, the sounds representing successive letters blend into each other, so the conversion of the analog signal to discrete characters can be a very difficult process.@@@@1@29@@danf@17-8-2009
10590440@unknown@formal@none@1@S@Also, in [[natural speech]] there are hardly any pauses between successive words; the location of those boundaries usually must take into account [[grammatical]] and [[semantic]] constraints, as well as the [[context]].@@@@1@31@@danf@17-8-2009
10590450@unknown@formal@none@1@S@; [[Text segmentation]]:@@@@1@3@@danf@17-8-2009
10590460@unknown@formal@none@1@S@Some written languages like [[Chinese language|Chinese]], [[Japanese language|Japanese]] and [[Thai language|Thai]] do not have single-word boundaries either, so any significant text [[parsing]] usually requires the identification of word boundaries, which is often a non-trivial task.@@@@1@35@@danf@17-8-2009
10590470@unknown@formal@none@1@S@; [[Word sense disambiguation]]:@@@@1@4@@danf@17-8-2009
10590480@unknown@formal@none@1@S@Many words have more than one [[meaning]]; we have to select the meaning which makes the most sense in context.@@@@1@20@@danf@17-8-2009
10590490@unknown@formal@none@1@S@; [[Syntactic ambiguity]]:@@@@1@3@@danf@17-8-2009
10590500@unknown@formal@none@1@S@The [[grammar]] for [[natural language]]s is [[ambiguous]], i.e. there are often multiple possible [[parse tree]]s for a given sentence.@@@@1@19@@danf@17-8-2009
10590510@unknown@formal@none@1@S@Choosing the most appropriate one usually requires [[semantics|semantic]] and contextual information.@@@@1@11@@danf@17-8-2009
10590520@unknown@formal@none@1@S@Specific problem components of syntactic ambiguity include [[sentence boundary disambiguation]].@@@@1@10@@danf@17-8-2009
10590530@unknown@formal@none@1@S@; Imperfect or irregular input :@@@@1@6@@danf@17-8-2009
10590540@unknown@formal@none@1@S@Foreign or regional accents and vocal impediments in speech; typing or grammatical errors, [[Optical character recognition|OCR]] errors in texts.@@@@1@19@@danf@17-8-2009
10590550@unknown@formal@none@1@S@; [[Speech acts]] and plans:@@@@1@5@@danf@17-8-2009
10590560@unknown@formal@none@1@S@A sentence can often be considered an action by the speaker.@@@@1@11@@danf@17-8-2009
10590570@unknown@formal@none@1@S@The sentence structure, alone, may not contain enough information to define this action.@@@@1@13@@danf@17-8-2009
10590580@unknown@formal@none@1@S@For instance, a question is actually the speaker requesting some sort of response from the listener.@@@@1@16@@danf@17-8-2009
10590590@unknown@formal@none@1@S@The desired response may be verbal, physical, or some combination.@@@@1@10@@danf@17-8-2009
10590600@unknown@formal@none@1@S@For example, "Can you pass the class?" is a request for a simple yes-or-no answer, while "Can you pass the salt?" is requesting a physical action to be performed.@@@@1@29@@danf@17-8-2009
10590610@unknown@formal@none@1@S@It is not appropriate to respond with "Yes, I can pass the salt," without the accompanying action (although "No" or "I can't reach the salt" would explain a lack of action).@@@@1@31@@danf@17-8-2009
10590620@unknown@formal@none@1@S@== Statistical NLP ==@@@@1@4@@danf@17-8-2009
10590630@unknown@formal@none@1@S@Statistical natural-language processing uses [[stochastic]], [[probabilistic]] and [[statistical]] methods to resolve some of the difficulties discussed above, especially those which arise because longer sentences are highly ambiguous when processed with realistic grammars, yielding thousands or millions of possible analyses.@@@@1@39@@danf@17-8-2009
10590640@unknown@formal@none@1@S@Methods for disambiguation often involve the use of [[corpus linguistics | corpora]] and [[Markov model]]s.@@@@1@15@@danf@17-8-2009
10590650@unknown@formal@none@1@S@Statistical NLP comprises all quantitative approaches to automated language processing, including probabilistic modeling, [[information theory]], and [[linear algebra]].@@@@1@18@@danf@17-8-2009
10590660@unknown@formal@none@1@S@The technology for statistical NLP comes mainly from [[machine learning]] and [[data mining]], both of which are fields of [[artificial intelligence]] that involve learning from data.@@@@1@26@@danf@17-8-2009
10590670@unknown@formal@none@1@S@==Major tasks in NLP==@@@@1@4@@danf@17-8-2009
10590680@unknown@formal@none@1@S@* [[Automatic summarization]]@@@@1@3@@danf@17-8-2009
10590690@unknown@formal@none@1@S@* [[Foreign language reading aid]]@@@@1@5@@danf@17-8-2009
10590700@unknown@formal@none@1@S@* [[Foreign language writing aid]]@@@@1@5@@danf@17-8-2009
10590710@unknown@formal@none@1@S@* [[Information extraction]]@@@@1@3@@danf@17-8-2009
10590720@unknown@formal@none@1@S@* [[Information retrieval]]@@@@1@3@@danf@17-8-2009
10590730@unknown@formal@none@1@S@* [[Machine translation]]@@@@1@3@@danf@17-8-2009
10590740@unknown@formal@none@1@S@* [[Named entity recognition]]@@@@1@4@@danf@17-8-2009
10590750@unknown@formal@none@1@S@* [[Natural language generation]]@@@@1@4@@danf@17-8-2009
10590760@unknown@formal@none@1@S@* [[Natural language understanding]]@@@@1@4@@danf@17-8-2009
10590770@unknown@formal@none@1@S@* [[Optical character recognition]]@@@@1@4@@danf@17-8-2009
10590780@unknown@formal@none@1@S@* [[Question answering]]@@@@1@3@@danf@17-8-2009
10590790@unknown@formal@none@1@S@* [[Speech recognition]]@@@@1@3@@danf@17-8-2009
10590800@unknown@formal@none@1@S@* [[Spoken dialogue system]]@@@@1@4@@danf@17-8-2009
10590810@unknown@formal@none@1@S@* [[Text simplification]]@@@@1@3@@danf@17-8-2009
10590820@unknown@formal@none@1@S@* [[Text to speech]]@@@@1@4@@danf@17-8-2009
10590830@unknown@formal@none@1@S@* [[Text-proofing]]@@@@1@2@@danf@17-8-2009
10590840@unknown@formal@none@1@S@== Evaluation of natural language processing ==@@@@1@7@@danf@17-8-2009
10590850@unknown@formal@none@1@S@===Objectives===@@@@1@1@@danf@17-8-2009
10590860@unknown@formal@none@1@S@The goal of NLP evaluation is to measure one or more ''qualities'' of an algorithm or a system, in order to determine if (or to what extent) the system answers the goals of its designers, or the needs of its users.@@@@1@41@@danf@17-8-2009
10590870@unknown@formal@none@1@S@Research in NLP evaluation has received considerable attention, because the definition of proper evaluation criteria is one way to specify precisely an NLP problem, going thus beyond the vagueness of tasks defined only as ''language understanding'' or ''language generation''.@@@@1@39@@danf@17-8-2009
10590880@unknown@formal@none@1@S@A precise set of evaluation criteria, which includes mainly evaluation data and evaluation metrics, enables several teams to compare their solutions to a given NLP problem.@@@@1@26@@danf@17-8-2009
10590890@unknown@formal@none@1@S@===Short history of evaluation in NLP===@@@@1@6@@danf@17-8-2009
10590900@unknown@formal@none@1@S@The first evaluation campaign on written texts seems to be a campaign dedicated to message understanding in 1987 (Pallet 1998).@@@@1@20@@danf@17-8-2009
10590910@unknown@formal@none@1@S@Then, the Parseval/GEIG project compared phrase-structure grammars (Black 1991).@@@@1@9@@danf@17-8-2009
10590920@unknown@formal@none@1@S@A series of campaigns within Tipster project were realized on tasks like summarization, translation and searching (Hirshman 1998).@@@@1@18@@danf@17-8-2009
10590930@unknown@formal@none@1@S@In 1994, in Germany, the Morpholympics compared German taggers.@@@@1@9@@danf@17-8-2009
10590940@unknown@formal@none@1@S@Then, the Senseval and Romanseval campaigns were conducted with the objectives of semantic disambiguation.@@@@1@14@@danf@17-8-2009
10590950@unknown@formal@none@1@S@In 1996, the Sparkle campaign compared syntactic parsers in four different languages (English, French, German and Italian).@@@@1@17@@danf@17-8-2009
10590960@unknown@formal@none@1@S@In France, the Grace project compared a set of 21 taggers for French in 1997 (Adda 1999).@@@@1@17@@danf@17-8-2009
10590970@unknown@formal@none@1@S@In 2004, during the [[Technolangue/Easy]] project, 13 parsers for French were compared.@@@@1@12@@danf@17-8-2009
10590980@unknown@formal@none@1@S@Large-scale evaluation of dependency parsers were performed in the context of the CoNLL shared tasks in 2006 and 2007.@@@@1@19@@danf@17-8-2009
10590990@unknown@formal@none@1@S@In Italy, the evalita campaign was conducted in 2007 to compare various tools for Italian [http://evalita.itc.it evalita web site].@@@@1@19@@danf@17-8-2009
10591000@unknown@formal@none@1@S@In France, within the ANR-Passage project (end of 2007), 10 parsers for French were compared [http://atoll.inria.fr/passage/ passage web site].@@@@1@19@@danf@17-8-2009
10591010@unknown@formal@none@1@S@Adda G., Mariani J., Paroubek P., Rajman M. 1999 L'action GRACE d'évaluation de l'assignation des parties du discours pour le français. Langues vol-2@@@@1@23@@danf@17-8-2009
10591030@unknown@formal@none@1@S@Black E., Abney S., Flickinger D., Gdaniec C., Grishman R., Harrison P., Hindle D., Ingria R., Jelinek F., Klavans J., Liberman M., Marcus M., Reukos S., Santoni B., Strzalkowski T. 1991 A procedure for quantitatively comparing the syntactic coverage of English grammars. DARPA Speech and Natural Language Workshop@@@@1@48@@danf@17-8-2009
10591050@unknown@formal@none@1@S@Hirshman L. 1998 Language understanding evaluation: lessons learned from MUC and ATIS. LREC Granada@@@@1@14@@danf@17-8-2009
10591070@unknown@formal@none@1@S@Pallet D.S. 1998 The NIST role in automatic speech recognition benchmark tests. LREC Granada@@@@1@14@@danf@17-8-2009
10591090@unknown@formal@none@1@S@===Different types of evaluation===@@@@1@4@@danf@17-8-2009
10591100@unknown@formal@none@1@S@Depending on the evaluation procedures, a number of distinctions are traditionally made in NLP evaluation.@@@@1@15@@danf@17-8-2009
10591110@unknown@formal@none@1@S@* Intrinsic vs. extrinsic evaluation@@@@1@5@@danf@17-8-2009
10591120@unknown@formal@none@1@S@Intrinsic evaluation considers an isolated NLP system and characterizes its performance mainly with respect to a ''gold standard'' result, pre-defined by the evaluators.@@@@1@23@@danf@17-8-2009
10591130@unknown@formal@none@1@S@Extrinsic evaluation, also called ''evaluation in use'' considers the NLP system in a more complex setting, either as an embedded system or serving a precise function for a human user.@@@@1@30@@danf@17-8-2009
10591140@unknown@formal@none@1@S@The extrinsic performance of the system is then characterized in terms of its utility with respect to the overall task of the complex system or the human user.@@@@1@28@@danf@17-8-2009
10591150@unknown@formal@none@1@S@* Black-box vs. glass-box evaluation@@@@1@5@@danf@17-8-2009
10591160@unknown@formal@none@1@S@Black-box evaluation requires one to run an NLP system on a given data set and to measure a number of parameters related to the quality of the process (speed, reliability, resource consumption) and, most importantly, to the quality of the result (e.g. the accuracy of data annotation or the fidelity of a translation).@@@@1@53@@danf@17-8-2009
10591170@unknown@formal@none@1@S@Glass-box evaluation looks at the design of the system, the algorithms that are implemented, the linguistic resources it uses (e.g. vocabulary size), etc.@@@@1@23@@danf@17-8-2009
10591180@unknown@formal@none@1@S@Given the complexity of NLP problems, it is often difficult to predict performance only on the basis of glass-box evaluation, but this type of evaluation is more informative with respect to error analysis or future developments of a system.@@@@1@39@@danf@17-8-2009
10591190@unknown@formal@none@1@S@* Automatic vs. manual evaluation@@@@1@5@@danf@17-8-2009
10591200@unknown@formal@none@1@S@In many cases, automatic procedures can be defined to evaluate an NLP system by comparing its output with the gold standard (or desired) one.@@@@1@24@@danf@17-8-2009
10591210@unknown@formal@none@1@S@Although the cost of producing the gold standard can be quite high, automatic evaluation can be repeated as often as needed without much additional costs (on the same input data).@@@@1@30@@danf@17-8-2009
10591220@unknown@formal@none@1@S@However, for many NLP problems, the definition of a gold standard is a complex task, and can prove impossible when inter-annotator agreement is insufficient.@@@@1@24@@danf@17-8-2009
10591230@unknown@formal@none@1@S@Manual evaluation is performed by human judges, which are instructed to estimate the quality of a system, or most often of a sample of its output, based on a number of criteria.@@@@1@32@@danf@17-8-2009
10591240@unknown@formal@none@1@S@Although, thanks to their linguistic competence, human judges can be considered as the reference for a number of language processing tasks, there is also considerable variation across their ratings.@@@@1@29@@danf@17-8-2009
10591250@unknown@formal@none@1@S@This is why automatic evaluation is sometimes referred to as ''objective'' evaluation, while the human kind appears to be more ''subjective.''@@@@1@21@@danf@17-8-2009
10591260@unknown@formal@none@1@S@=== Shared tasks (Campaigns)===@@@@1@4@@danf@17-8-2009
10591270@unknown@formal@none@1@S@* [[BioCreative]]@@@@1@2@@danf@17-8-2009
10591280@unknown@formal@none@1@S@* [[Message Understanding Conference]]@@@@1@4@@danf@17-8-2009
10591290@unknown@formal@none@1@S@* [[Technolangue/Easy]]@@@@1@2@@danf@17-8-2009
10591300@unknown@formal@none@1@S@* [[Text Retrieval Conference]]@@@@1@4@@danf@17-8-2009
10591310@unknown@formal@none@1@S@==Standardization in NLP==@@@@1@3@@danf@17-8-2009
10591320@unknown@formal@none@1@S@An ISO sub-committee is working in order to ease interoperability between [[Lexical resource]]s and NLP programs.@@@@1@16@@danf@17-8-2009
10591330@unknown@formal@none@1@S@The sub-committee is part of [[ISO/TC37]] and is called ISO/TC37/SC4.@@@@1@10@@danf@17-8-2009
10591340@unknown@formal@none@1@S@Some ISO standards are already published but most of them are under construction, mainly on lexicon representation (see [[lexical markup framework|LMF]]), annotation and data category registry.@@@@1@26@@danf@17-8-2009
10600010@unknown@formal@none@1@S@Neural network@@@@1@2@@danf@17-8-2009
10600020@unknown@formal@none@1@S@Traditionally, the term '''neural network''' had been used to refer to a network or circuit of [[neuron|biological neurons]].@@@@1@18@@danf@17-8-2009
10600030@unknown@formal@none@1@S@The modern usage of the term often refers to [[artificial neural network]]s, which are composed of [[artificial neuron]]s or nodes.@@@@1@20@@danf@17-8-2009
10600040@unknown@formal@none@1@S@Thus the term has two distinct usages:@@@@1@7@@danf@17-8-2009
10600050@unknown@formal@none@1@S@# [[Biological neural network]]s are made up of real biological neurons that are connected or functionally-related in the [[peripheral nervous system]] or the [[central nervous system]].@@@@1@26@@danf@17-8-2009
10600060@unknown@formal@none@1@S@In the field of [[neuroscience]], they are often identified as groups of neurons that perform a specific physiological function in laboratory analysis.@@@@1@22@@danf@17-8-2009
10600070@unknown@formal@none@1@S@# [[Artificial neural network]]s are made up of interconnecting artificial neurons (programming constructs that mimic the properties of biological neurons).@@@@1@20@@danf@17-8-2009
10600080@unknown@formal@none@1@S@Artificial neural networks may either be used to gain an understanding of biological neural networks, or for solving artificial intelligence problems without necessarily creating a model of a real biological system.@@@@1@31@@danf@17-8-2009
10600090@unknown@formal@none@1@S@This article focuses on the relationship between the two concepts; for detailed coverage of the two different concepts refer to the separate articles: [[Biological neural network]] and [[Artificial neural network]].@@@@1@30@@danf@17-8-2009
10600100@unknown@formal@none@1@S@==Characterization==@@@@1@1@@danf@17-8-2009
10600110@unknown@formal@none@1@S@In general a biological neural network is composed of a group or groups of chemically connected or functionally associated neurons.@@@@1@20@@danf@17-8-2009
10600120@unknown@formal@none@1@S@A single neuron may be connected to many other neurons and the total number of neurons and connections in a network may be extensive.@@@@1@24@@danf@17-8-2009
10600130@unknown@formal@none@1@S@Connections, called [[synapses]], are usually formed from [[axons]] to [[dendrites]], though dendrodendritic microcircuits and other connections are possible.@@@@1@18@@danf@17-8-2009
10600140@unknown@formal@none@1@S@Apart from the electrical signaling, there are other forms of signaling that arise from [[neurotransmitter]] diffusion, which have an effect on electrical signaling.@@@@1@23@@danf@17-8-2009
10600150@unknown@formal@none@1@S@As such, neural networks are extremely complex.@@@@1@7@@danf@17-8-2009
10600160@unknown@formal@none@1@S@[[Artificial intelligence]] and [[cognitive modeling]] try to simulate some properties of neural networks.@@@@1@13@@danf@17-8-2009
10600170@unknown@formal@none@1@S@While similar in their techniques, the former has the aim of solving particular tasks, while the latter aims to build mathematical models of biological neural systems.@@@@1@26@@danf@17-8-2009
10600180@unknown@formal@none@1@S@In the [[artificial intelligence]] field, artificial neural networks have been applied successfully to [[speech recognition]], [[image analysis]] and adaptive [[control]], in order to construct [[software agents]] (in [[Video game|computer and video games]]) or [[autonomous robot]]s.@@@@1@35@@danf@17-8-2009
10600190@unknown@formal@none@1@S@Most of the currently employed artificial neural networks for artificial intelligence are based on [[statistical estimation]], [[Optimization (mathematics)|optimization]] and [[control theory]].@@@@1@21@@danf@17-8-2009
10600200@unknown@formal@none@1@S@The [[cognitive modelling]] field involves the physical or mathematical modeling of the behaviour of neural systems; ranging from the individual neural level (e.g. modelling the spike response curves of neurons to a stimulus), through the neural cluster level (e.g. modelling the release and effects of dopamine in the basal ganglia) to the complete organism (e.g. behavioural modelling of the organism's response to stimuli).@@@@1@63@@danf@17-8-2009
10600210@unknown@formal@none@1@S@==The brain, neural networks and computers==@@@@1@6@@danf@17-8-2009
10600220@unknown@formal@none@1@S@Neural networks, as used in artificial intelligence, have traditionally been viewed as simplified models of neural processing in the brain, even though the relation between this model and brain biological architecture is debated.@@@@1@33@@danf@17-8-2009
10600230@unknown@formal@none@1@S@A subject of current research in theoretical neuroscience is the question surrounding the degree of complexity and the properties that individual neural elements should have to reproduce something resembling animal intelligence.@@@@1@31@@danf@17-8-2009
10600240@unknown@formal@none@1@S@Historically, computers evolved from the [[von Neumann architecture]], which is based on sequential processing and execution of explicit instructions.@@@@1@19@@danf@17-8-2009
10600250@unknown@formal@none@1@S@On the other hand, the origins of neural networks are based on efforts to model information processing in biological systems, which may rely largely on parallel processing as well as implicit instructions based on recognition of patterns of 'sensory' input from external sources.@@@@1@43@@danf@17-8-2009
10600260@unknown@formal@none@1@S@In other words, at its very heart a neural network is a complex statistical processor (as opposed to being tasked to sequentially process and execute).@@@@1@25@@danf@17-8-2009
10600270@unknown@formal@none@1@S@==Neural networks and artificial intelligence==@@@@1@5@@danf@17-8-2009
10600280@unknown@formal@none@1@S@An ''artificial neural network'' (ANN), also called a ''simulated neural network'' (SNN) or commonly just ''neural network'' (NN) is an interconnected group of [[artificial neuron]]s that uses a [[mathematical model|mathematical or computational model]] for [[information processing]] based on a [[connectionism|connectionistic]] approach to [[computation]].@@@@1@43@@danf@17-8-2009
10600290@unknown@formal@none@1@S@In most cases an ANN is an [[adaptive system]] that changes its structure based on external or internal information that flows through the network.@@@@1@24@@danf@17-8-2009
10600300@unknown@formal@none@1@S@In more practical terms neural networks are [[non-linear]] [[statistical]] [[data modeling]] or [[decision making]] tools.@@@@1@15@@danf@17-8-2009
10600310@unknown@formal@none@1@S@They can be used to model complex relationships between inputs and outputs or to [[Pattern recognition|find patterns]] in data.@@@@1@19@@danf@17-8-2009
10600320@unknown@formal@none@1@S@===Background===@@@@1@1@@danf@17-8-2009
10600330@unknown@formal@none@1@S@An [[artificial neural network]] involves a network of simple processing elements ([[artificial neurons]]) which can exhibit complex global behaviour, determined by the connections between the processing elements and element parameters.@@@@1@30@@danf@17-8-2009
10600340@unknown@formal@none@1@S@One classical type of artificial neural network is the [[Hopfield net]].@@@@1@11@@danf@17-8-2009
10600350@unknown@formal@none@1@S@In a neural network model simple [[Node (neural networks)|nodes]], which can be called variously "neurons", "neurodes", "Processing Elements" (PE) or "units", are connected together to form a network of nodes — hence the term "neural network".@@@@1@36@@danf@17-8-2009
10600360@unknown@formal@none@1@S@While a neural network does not have to be adaptive ''per se'', its practical use comes with algorithms designed to alter the strength (weights) of the connections in the network to produce a desired signal flow.@@@@1@36@@danf@17-8-2009
10600370@unknown@formal@none@1@S@In modern [[Neural network software|software implementations]] of artificial neural networks the approach inspired by biology has more or less been abandoned for a more practical approach based on statistics and signal processing.@@@@1@32@@danf@17-8-2009
10600380@unknown@formal@none@1@S@In some of these systems neural networks, or parts of neural networks (such as [[artificial neuron]]s) are used as components in larger systems that combine both adaptive and non-adaptive elements.@@@@1@30@@danf@17-8-2009
10600390@unknown@formal@none@1@S@The concept of a neural network appears to have first been proposed by [[Alan Turing]] in his 1948 paper "Intelligent Machinery".@@@@1@21@@danf@17-8-2009
10600400@unknown@formal@none@1@S@===Applications===@@@@1@1@@danf@17-8-2009
10600410@unknown@formal@none@1@S@The utility of artificial neural network models lies in the fact that they can be used to infer a function from observations and also to use it.@@@@1@27@@danf@17-8-2009
10600420@unknown@formal@none@1@S@This is particularly useful in applications where the complexity of the data or task makes the design of such a function by hand impractical.@@@@1@24@@danf@17-8-2009
10600430@unknown@formal@none@1@S@====Real life applications====@@@@1@3@@danf@17-8-2009
10600440@unknown@formal@none@1@S@The tasks to which artificial neural networks are applied tend to fall within the following broad categories:@@@@1@17@@danf@17-8-2009
10600450@unknown@formal@none@1@S@*[[Function approximation]], or [[regression analysis]], including [[time series prediction]] and modelling.@@@@1@11@@danf@17-8-2009
10600460@unknown@formal@none@1@S@*[[Statistical classification|Classification]], including [[Pattern recognition|pattern]] and sequence recognition, novelty detection and sequential decision making.@@@@1@14@@danf@17-8-2009
10600470@unknown@formal@none@1@S@*[[Data processing]], including filtering, clustering, [[blind signal separation]] and compression.@@@@1@10@@danf@17-8-2009
10600480@unknown@formal@none@1@S@Application areas include system identification and control (vehicle control, process control), game-playing and decision making (backgammon, chess, racing), pattern recognition (radar systems, face identification, object recognition, etc.), sequence recognition (gesture, speech, handwritten text recognition), medical diagnosis, financial applications, [[data mining]] (or knowledge discovery in databases, "KDD"), visualization and [[e-mail spam]] filtering.@@@@1@51@@danf@17-8-2009
10600490@unknown@formal@none@1@S@===Neural network software===@@@@1@3@@danf@17-8-2009
10600500@unknown@formal@none@1@S@''Main article:'' [[Neural network software]]@@@@1@5@@danf@17-8-2009
10600510@unknown@formal@none@1@S@'''Neural network software''' is used to [[Simulation|simulate]], [[research]], [[Software development|develop]] and apply [[artificial neural network]]s, [[biological neural network]]s and in some cases a wider array of [[adaptive system]]s.@@@@1@28@@danf@17-8-2009
10600520@unknown@formal@none@1@S@====Learning paradigms====@@@@1@2@@danf@17-8-2009
10600530@unknown@formal@none@1@S@There are three major learning paradigms, each corresponding to a particular abstract learning task.@@@@1@14@@danf@17-8-2009
10600540@unknown@formal@none@1@S@These are [[supervised learning]], [[unsupervised learning]] and [[reinforcement learning]].@@@@1@9@@danf@17-8-2009
10600550@unknown@formal@none@1@S@Usually any given type of network architecture can be employed in any of those tasks.@@@@1@15@@danf@17-8-2009
10600560@unknown@formal@none@1@S@;Supervised learning@@@@1@2@@danf@17-8-2009
10600570@unknown@formal@none@1@S@In [[supervised learning]], we are given a set of example pairs and the aim is to find a function in the allowed class of functions that matches the examples.@@@@1@39@@danf@17-8-2009
10600580@unknown@formal@none@1@S@In other words, we wish to ''infer'' how the mapping implied by the data and the cost function is related to the mismatch between our mapping and the data.@@@@1@29@@danf@17-8-2009
10600590@unknown@formal@none@1@S@;Unsupervised learning@@@@1@2@@danf@17-8-2009
10600600@unknown@formal@none@1@S@In [[unsupervised learning]] we are given some data , and a cost function which is to be minimized which can be any function of and the network's output, .@@@@1@30@@danf@17-8-2009
10600610@unknown@formal@none@1@S@The cost function is determined by the task formulation.@@@@1@9@@danf@17-8-2009
10600620@unknown@formal@none@1@S@Most applications fall within the domain of [[estimation problems]] such as [[statistical modeling]], [[Data compression|compression]], [[Mail filter|filtering]], [[blind source separation]] and [[data clustering|clustering]].@@@@1@23@@danf@17-8-2009
10600630@unknown@formal@none@1@S@;Reinforcement learning@@@@1@2@@danf@17-8-2009
10600640@unknown@formal@none@1@S@In [[reinforcement learning]], data is usually not given, but generated by an agent's interactions with the environment.@@@@1@18@@danf@17-8-2009
10600650@unknown@formal@none@1@S@At each point in time , the agent performs an action and the environment generates an observation and an instantaneous cost , according to some (usually unknown) dynamics.@@@@1@30@@danf@17-8-2009
10600660@unknown@formal@none@1@S@The aim is to discover a ''policy'' for selecting actions that minimises some measure of a long-term cost, i.e. the expected cumulative cost.@@@@1@23@@danf@17-8-2009
10600670@unknown@formal@none@1@S@The environment's dynamics and the long-term cost for each policy are usually unknown, but can be estimated.@@@@1@17@@danf@17-8-2009
10600680@unknown@formal@none@1@S@ANNs are frequently used in reinforcement learning as part of the overall algorithm.@@@@1@13@@danf@17-8-2009
10600690@unknown@formal@none@1@S@Tasks that fall within the paradigm of reinforcement learning are [[control]] problems, [[game]]s and other [[sequential decision making]] tasks.@@@@1@19@@danf@17-8-2009
10600700@unknown@formal@none@1@S@====Learning algorithms====@@@@1@2@@danf@17-8-2009
10600710@unknown@formal@none@1@S@There are many algorithms for training neural networks; most of them can be viewed as a straightforward application of [[Optimization (mathematics)|optimization]] theory and [[statistical estimation]].@@@@1@25@@danf@17-8-2009
10600720@unknown@formal@none@1@S@[[Evolutionary computation]] methods, [[simulated annealing]], [[Expectation-Maximization|expectation maximization]] and [[non-parametric methods]] are among other commonly used methods for training neural networks.@@@@1@20@@danf@17-8-2009
10600730@unknown@formal@none@1@S@See also [[machine learning]].@@@@1@4@@danf@17-8-2009
10600740@unknown@formal@none@1@S@Recent developments in this field also saw the use of [[particle swarm optimization]] and other [[swarm intelligence]] techniques used in the training of neural networks.@@@@1@25@@danf@17-8-2009
10600750@unknown@formal@none@1@S@==Neural networks and neuroscience==@@@@1@4@@danf@17-8-2009
10600760@unknown@formal@none@1@S@Theoretical and [[computational neuroscience]] is the field concerned with the theoretical analysis and computational modeling of biological neural systems.@@@@1@19@@danf@17-8-2009
10600770@unknown@formal@none@1@S@Since neural systems are intimately related to cognitive processes and behaviour, the field is closely related to cognitive and behavioural modeling.@@@@1@21@@danf@17-8-2009
10600780@unknown@formal@none@1@S@The aim of the field is to create models of biological neural systems in order to understand how biological systems work.@@@@1@21@@danf@17-8-2009
10600790@unknown@formal@none@1@S@To gain this understanding, neuroscientists strive to make a link between observed biological processes (data), biologically plausible mechanisms for neural processing and learning ([[biological neural network]] models) and theory (statistical learning theory and [[information theory]]).@@@@1@35@@danf@17-8-2009
10600800@unknown@formal@none@1@S@=== Types of models ===@@@@1@5@@danf@17-8-2009
10600810@unknown@formal@none@1@S@Many models are used in the field, each defined at a different level of abstraction and trying to model different aspects of neural systems.@@@@1@24@@danf@17-8-2009
10600820@unknown@formal@none@1@S@They range from models of the short-term behaviour of [[biological neuron models|individual neurons]], through models of how the dynamics of neural circuitry arise from interactions between individual neurons, to models of how behaviour can arise from abstract neural modules that represent complete subsystems.@@@@1@43@@danf@17-8-2009
10600830@unknown@formal@none@1@S@These include models of the long-term and short-term plasticity of neural systems and its relation to learning and memory, from the individual neuron to the system level.@@@@1@27@@danf@17-8-2009
10600840@unknown@formal@none@1@S@===Current research===@@@@1@2@@danf@17-8-2009
10600850@unknown@formal@none@1@S@While initially research had been concerned mostly with the electrical characteristics of neurons, a particularly important part of the investigation in recent years has been the exploration of the role of [[neuromodulators]] such as [[dopamine]], [[acetylcholine]], and [[serotonin]] on behaviour and learning.@@@@1@42@@danf@17-8-2009
10600860@unknown@formal@none@1@S@[[Biophysics|Biophysical]] models, such as [[BCM theory]], have been important in understanding mechanisms for [[synaptic plasticity]], and have had applications in both computer science and neuroscience.@@@@1@25@@danf@17-8-2009
10600870@unknown@formal@none@1@S@Research is ongoing in understanding the computational algorithms used in the brain, with some recent biological evidence for [[radial basis networks]] and [[neural backpropagation]] as mechanisms for processing data.@@@@1@29@@danf@17-8-2009
10600880@unknown@formal@none@1@S@==History of the neural network analogy==@@@@1@6@@danf@17-8-2009
10600890@unknown@formal@none@1@S@The concept of neural networks started in the late-1800s as an effort to describe how the human mind performed.@@@@1@19@@danf@17-8-2009
10600900@unknown@formal@none@1@S@These ideas started being applied to computational models with the [[Perceptron]].@@@@1@11@@danf@17-8-2009
10600910@unknown@formal@none@1@S@In early 1950s [[Friedrich Hayek]] was one of the first to posit the idea of [[spontaneous order]] in the brain arising out of decentralized networks of simple units (neurons).@@@@1@29@@danf@17-8-2009
10600920@unknown@formal@none@1@S@In the late 1940s, [[Donald Hebb]] made one of the first hypotheses for a mechanism of neural plasticity (i.e. learning), [[Hebbian learning]].@@@@1@22@@danf@17-8-2009
10600930@unknown@formal@none@1@S@Hebbian learning is considered to be a 'typical' unsupervised learning rule and it (and variants of it) was an early model for [[long term potentiation]].@@@@1@25@@danf@17-8-2009
10600940@unknown@formal@none@1@S@The [[Perceptron]] is essentially a linear classifier for classifying data specified by parameters and an output function .@@@@1@32@@danf@17-8-2009
10600950@unknown@formal@none@1@S@Its parameters are adapted with an ad-hoc rule similar to stochastic steepest gradient descent.@@@@1@14@@danf@17-8-2009
10600960@unknown@formal@none@1@S@Because the [[inner product]] is a [[linear operator]] in the input space, the Perceptron can only perfectly classify a set of data for which different classes are [[linearly separable]] in the input space, while it often fails completely for non-separable data.@@@@1@41@@danf@17-8-2009
10600970@unknown@formal@none@1@S@While the development of the algorithm initially generated some enthusiasm, partly because of its apparent relation to biological mechanisms, the later discovery of this inadequacy caused such models to be abandoned until the introduction of non-linear models into the field.@@@@1@40@@danf@17-8-2009
10600980@unknown@formal@none@1@S@The [[Cognitron]] (1975) was an early multilayered neural network with a training algorithm.@@@@1@13@@danf@17-8-2009
10600990@unknown@formal@none@1@S@The actual structure of the network and the methods used to set the interconnection weights change from one neural strategy to another, each with its advantages and disadvantages.@@@@1@28@@danf@17-8-2009
10601000@unknown@formal@none@1@S@Networks can propagate information in one direction only, or they can bounce back and forth until self-activation at a node occurs and the network settles on a final state.@@@@1@29@@danf@17-8-2009
10601010@unknown@formal@none@1@S@The ability for bi-directional flow of inputs between neurons/nodes was produced with the [[Hopfield net|Hopfield's network]] (1982), and specialization of these node layers for specific purposes was introduced through the first [[hybrid neural network|hybrid network]].@@@@1@35@@danf@17-8-2009
10601020@unknown@formal@none@1@S@The [[connectionism|parallel distributed processing]] of the mid-1980s became popular under the name [[connectionism]].@@@@1@13@@danf@17-8-2009
10601030@unknown@formal@none@1@S@The rediscovery of the [[backpropagation]] algorithm was probably the main reason behind the repopularisation of neural networks after the publication of "Learning Internal Representations by Error Propagation" in 1986 (Though backpropagation itself dates from 1974).@@@@1@35@@danf@17-8-2009
10601040@unknown@formal@none@1@S@The original network utilised multiple layers of weight-sum units of the type , where was a [[sigmoid function]] or [[logistic function]] such as used in [[logistic regression]].@@@@1@32@@danf@17-8-2009
10601050@unknown@formal@none@1@S@Training was done by a form of stochastic steepest gradient descent.@@@@1@11@@danf@17-8-2009
10601060@unknown@formal@none@1@S@The employment of the chain rule of differentiation in deriving the appropriate parameter updates results in an algorithm that seems to 'backpropagate errors', hence the nomenclature.@@@@1@26@@danf@17-8-2009
10601070@unknown@formal@none@1@S@However it is essentially a form of gradient descent.@@@@1@9@@danf@17-8-2009
10601080@unknown@formal@none@1@S@Determining the optimal parameters in a model of this type is not trivial, and steepest gradient descent methods cannot be relied upon to give the solution without a good starting point.@@@@1@31@@danf@17-8-2009
10601090@unknown@formal@none@1@S@In recent times, networks with the same architecture as the backpropagation network are referred to as [[Multilayer perceptron|Multi-Layer Perceptrons]].@@@@1@19@@danf@17-8-2009
10601100@unknown@formal@none@1@S@This name does not impose any limitations on the type of algorithm used for learning.@@@@1@15@@danf@17-8-2009
10601110@unknown@formal@none@1@S@The backpropagation network generated much enthusiasm at the time and there was much controversy about whether such learning could be implemented in the brain or not, partly because a mechanism for reverse signalling was not obvious at the time, but most importantly because there was no plausible source for the 'teaching' or 'target' signal.@@@@1@54@@danf@17-8-2009
10601120@unknown@formal@none@1@S@==Criticism==@@@@1@1@@danf@17-8-2009
10601130@unknown@formal@none@1@S@[[A. K. Dewdney]], a former ''[[Scientific American]]'' columnist, wrote in 1997, ''“Although neural nets do solve a few toy problems, their powers of computation are so limited that I am surprised anyone takes them seriously as a general problem-solving tool.”''@@@@1@40@@danf@17-8-2009
10601140@unknown@formal@none@1@S@(Dewdney, p.82)@@@@1@2@@danf@17-8-2009
10601150@unknown@formal@none@1@S@Arguments against Dewdney's position are that neural nets have been successfully used to solve many complex and diverse tasks, ranging from autonomously flying aircraft[http://www.nasa.gov/centers/dryden/news/NewsReleases/2003/03-49.html] to detecting credit card fraud[http://www.visa.ca/en/about/visabenefits/innovation.cfm].@@@@1@29@@danf@17-8-2009
10601160@unknown@formal@none@1@S@Technology writer [[Roger Bridgman]] commented on Dewdney's statements about neural nets:@@@@1@11@@danf@17-8-2009
10601170@unknown@formal@none@1@S@ Neural networks, for instance, are in the dock not only because they have been hyped to high heaven, (what hasn't?) but also because you could create a successful net without understanding how it worked: the bunch of numbers that captures its behaviour would in all probability be "an opaque, unreadable table...valueless as a scientific resource".@@@@1@56@@danf@17-8-2009
10601180@unknown@formal@none@1@S@In spite of his emphatic declaration that science is not technology, Dewdney seems here to pillory neural nets as bad science when most of those devising them are just trying to be good engineers.@@@@1@34@@danf@17-8-2009
10601190@unknown@formal@none@1@S@An unreadable table that a useful machine could read would still be well worth having.
@@@@1@15@@danf@17-8-2009
10610010@unknown@formal@none@1@S@N-gram@@@@1@1@@danf@17-8-2009
10610020@unknown@formal@none@1@S@An '''''n''-gram''' is a sub-sequence of ''n'' items from a given [[sequence]].@@@@1@12@@danf@17-8-2009
10610025@unknown@formal@none@1@S@''n''-grams are used in various areas of statistical [[natural language processing]] and genetic sequence analysis.@@@@1@15@@danf@17-8-2009
10610030@unknown@formal@none@1@S@The items in question can be letters, words or [[base pairs]] according to the application.@@@@1@15@@danf@17-8-2009
10610040@unknown@formal@none@1@S@An ''n''-gram of size 1 is a "[[unigram]]"; size 2 is a "[[bigram]]" (or, more etymologically sound but less commonly used, a "digram"); size 3 is a "[[trigram]]"; and size 4 or more is simply called an "''n''-gram".@@@@1@38@@danf@17-8-2009
10610050@unknown@formal@none@1@S@Some [[language model]]s built from n-grams are "(''n'' − 1)-order [[Markov_chain|Markov model]]s".@@@@1@10@@danf@17-8-2009
10610060@unknown@formal@none@1@S@==Examples==@@@@1@1@@danf@17-8-2009
10610070@unknown@formal@none@1@S@Here are examples of '''''word''''' level 3-grams and 4-grams (and counts of the number of times they appeared) from the [[N-gram#Google_use_of_N-gram|Google n-gram corpus]].@@@@1@23@@danf@17-8-2009
10610080@unknown@formal@none@1@S@*ceramics collectables collectibles (55)@@@@1@4@@danf@17-8-2009
10610090@unknown@formal@none@1@S@*ceramics collectables fine (130)@@@@1@4@@danf@17-8-2009
10610100@unknown@formal@none@1@S@*ceramics collected by (52)@@@@1@4@@danf@17-8-2009
10610110@unknown@formal@none@1@S@*ceramics collectible pottery (50)@@@@1@4@@danf@17-8-2009
10610120@unknown@formal@none@1@S@*ceramics collectibles cooking (45)@@@@1@4@@danf@17-8-2009
10610130@unknown@formal@none@1@S@4-grams@@@@1@1@@danf@17-8-2009
10610140@unknown@formal@none@1@S@*serve as the incoming (92)@@@@1@5@@danf@17-8-2009
10610150@unknown@formal@none@1@S@*serve as the incubator (99)@@@@1@5@@danf@17-8-2009
10610160@unknown@formal@none@1@S@*serve as the independent (794)@@@@1@5@@danf@17-8-2009
10610170@unknown@formal@none@1@S@*serve as the index (223)@@@@1@5@@danf@17-8-2009
10610180@unknown@formal@none@1@S@*serve as the indication (72)@@@@1@5@@danf@17-8-2009
10610190@unknown@formal@none@1@S@*serve as the indicator (120)@@@@1@5@@danf@17-8-2009
10610200@unknown@formal@none@1@S@==''n''-gram models==@@@@1@2@@danf@17-8-2009
10610210@unknown@formal@none@1@S@An '''''n''-gram model''' models sequences, notably natural languages, using the statistical properties of ''n''-grams.@@@@1@14@@danf@17-8-2009
10610220@unknown@formal@none@1@S@This idea can be traced to an experiment by [[Claude Shannon]]'s work in [[information theory]].@@@@1@15@@danf@17-8-2009
10610230@unknown@formal@none@1@S@His question was, given a sequence of letters (for example, the sequence "for ex"), what is the [[likelihood]] of the next letter?@@@@1@22@@danf@17-8-2009
10610240@unknown@formal@none@1@S@From training data, one can derive a [[probability distribution]] for the next letter given a history of size : ''a'' = 0.4, ''b'' = 0.00001, ''c'' = 0, ....; where the probabilities of all possible "next-letters" sum to 1.0.@@@@1@39@@danf@17-8-2009
10610250@unknown@formal@none@1@S@More concisely, an ''n''-gram model predicts based on .@@@@1@13@@danf@17-8-2009
10610260@unknown@formal@none@1@S@In Probability terms, this is nothing but .@@@@1@13@@danf@17-8-2009
10610270@unknown@formal@none@1@S@When used for [[language model|language modeling]] independence assumptions are made so that each word depends only on the last ''n'' words.@@@@1@21@@danf@17-8-2009
10610280@unknown@formal@none@1@S@This [[Markov model]] is used as an approximation of the true underlying language.@@@@1@13@@danf@17-8-2009
10610290@unknown@formal@none@1@S@This assumption is important because it massively simplifies the problem of learning the language model from data.@@@@1@17@@danf@17-8-2009
10610300@unknown@formal@none@1@S@In addition, because of the open nature of language, it is common to group words unknown to the language model together.@@@@1@21@@danf@17-8-2009
10610310@unknown@formal@none@1@S@''n''-gram models are widely used in statistical [[natural language processing]].@@@@1@10@@danf@17-8-2009
10610320@unknown@formal@none@1@S@In [[speech recognition]], [[phonemes]] and sequences of phonemes are modeled using a ''n''-gram distribution.@@@@1@14@@danf@17-8-2009
10610330@unknown@formal@none@1@S@For parsing, words are modeled such that each ''n''-gram is composed of ''n'' words.@@@@1@14@@danf@17-8-2009
10610340@unknown@formal@none@1@S@For [[language recognition]], sequences of letters are modeled for different languages.@@@@1@11@@danf@17-8-2009
10610350@unknown@formal@none@1@S@For a sequence of words, (for example "the dog smelled like a skunk"), the trigrams would be: "the dog smelled", "dog smelled like", "smelled like a", and "like a skunk".@@@@1@30@@danf@17-8-2009
10610360@unknown@formal@none@1@S@For sequences of characters, the 3-grams (sometimes referred to as "trigrams") that can be generated from "good morning" are "goo", "ood", "od ", "d m", " mo", "mor" and so forth.@@@@1@31@@danf@17-8-2009
10610370@unknown@formal@none@1@S@Some practitioners preprocess strings to remove spaces, most simply collapse whitespace to a single space while preserving paragraph marks.@@@@1@19@@danf@17-8-2009
10610380@unknown@formal@none@1@S@Punctuation is also commonly reduced or removed by preprocessing.@@@@1@9@@danf@17-8-2009
10610385@unknown@formal@none@1@S@''n''-grams can also be used for sequences of words or, in fact, for almost any type of data.@@@@1@18@@danf@17-8-2009
10610390@unknown@formal@none@1@S@They have been used for example for extracting features for clustering large sets of satellite earth images and for determining what part of the Earth a particular image came from.@@@@1@30@@danf@17-8-2009
10610400@unknown@formal@none@1@S@They have also been very successful as the first pass in genetic sequence search and in the identification of which species short sequences of DNA were taken from.@@@@1@28@@danf@17-8-2009
10610410@unknown@formal@none@1@S@N-gram models are often criticized because they lack any explicit representation of long range dependency.@@@@1@15@@danf@17-8-2009
10610420@unknown@formal@none@1@S@While it is true that the only explicit dependency range is (n-1) tokens for an n-gram model, it is also true that the effective range of dependency is significantly longer than this although long range correlations drop exponentially with distance for any Markov model.@@@@1@44@@danf@17-8-2009
10610430@unknown@formal@none@1@S@Alternative Markov language models that incorporate some degree of local state can exhibit very long range dependencies.@@@@1@17@@danf@17-8-2009
10610440@unknown@formal@none@1@S@This is often done using hand-crafted state variables that represent, for instance, the position in a sentence, the general topic of discourse or a grammatical state variable.@@@@1@27@@danf@17-8-2009
10610450@unknown@formal@none@1@S@Some of the best parsers of English currently in existence are roughly of this form.@@@@1@15@@danf@17-8-2009
10610460@unknown@formal@none@1@S@Another criticism that has been leveled is that Markov models of language, including n-gram models, do not explicitly capture the performance/competence distinction introduced by [[Noam Chomsky]].@@@@1@26@@danf@17-8-2009
10610470@unknown@formal@none@1@S@This criticism fails to explain why parsers that are the best at parsing text seem to uniformly lack any such distinction and most even lack any clear distinction between semantics and syntax.@@@@1@32@@danf@17-8-2009
10610480@unknown@formal@none@1@S@Most proponents of n-gram and related language models opt for a fairly pragmatic approach to language modeling that emphasizes empirical results over theoretical purity.@@@@1@24@@danf@17-8-2009
10610490@unknown@formal@none@1@S@==''n''-grams for approximate matching==@@@@1@4@@danf@17-8-2009
10610500@unknown@formal@none@1@S@''n''-grams can also be used for efficient approximate matching.@@@@1@9@@danf@17-8-2009
10610510@unknown@formal@none@1@S@By converting a sequence of items to a set of ''n''-grams, it can be embedded in a [[vector space]] (in other words, represented as a [[histogram]]), thus allowing the sequence to be compared to other sequences in an efficient manner.@@@@1@40@@danf@17-8-2009
10610520@unknown@formal@none@1@S@For example, if we convert strings with only letters in the English alphabet into 3-grams, we get a -dimensional space (the first dimension measures the number of occurrences of "aaa", the second "aab", and so forth for all possible combinations of three letters).@@@@1@43@@danf@17-8-2009
10610530@unknown@formal@none@1@S@Using this representation, we lose information about the string.@@@@1@9@@danf@17-8-2009
10610540@unknown@formal@none@1@S@For example, both the strings "abcba" and "bcbab" give rise to exactly the same 2-grams.@@@@1@15@@danf@17-8-2009
10610550@unknown@formal@none@1@S@However, we know empirically that if two strings of real text have a similar vector representation (as measured by [[dot product|cosine distance]]) then they are likely to be similar.@@@@1@29@@danf@17-8-2009
10610560@unknown@formal@none@1@S@Other metrics have also been applied to vectors of ''n''-grams with varying, sometimes better, results.@@@@1@15@@danf@17-8-2009
10610570@unknown@formal@none@1@S@For example [[z-score]]s have been used to compare documents by examining how many standard deviations each ''n''-gram differs from its mean occurrence in a large collection, or [[text corpus]], of documents (which form the "background" vector).@@@@1@36@@danf@17-8-2009
10610580@unknown@formal@none@1@S@In the event of small counts, the [[g-score]] may give better results for comparing alternative models.@@@@1@16@@danf@17-8-2009
10610590@unknown@formal@none@1@S@It is also possible to take a more principled approach to the statistics of ''n''-grams, modeling similarity as the likelihood that two strings came from the same source directly in terms of a problem in [[Bayesian inference]].@@@@1@37@@danf@17-8-2009
10610600@unknown@formal@none@1@S@==Other applications==@@@@1@2@@danf@17-8-2009
10610610@unknown@formal@none@1@S@''n''-grams find use in several areas of computer science, [[computational linguistics]], and applied mathematics.@@@@1@14@@danf@17-8-2009
10610620@unknown@formal@none@1@S@They have been used to:@@@@1@5@@danf@17-8-2009
10610630@unknown@formal@none@1@S@* design [[kernel (mathematics)|kernels]] that allow [[machine learning]] algorithms such as [[support vector machine]]s to learn from string data@@@@1@19@@danf@17-8-2009
10610640@unknown@formal@none@1@S@* find likely candidates for the correct spelling of a misspelled word@@@@1@12@@danf@17-8-2009
10610650@unknown@formal@none@1@S@* improve compression in [[data compression|compression algorithms]] where a small area of data requires ''n''-grams of greater length@@@@1@18@@danf@17-8-2009
10610660@unknown@formal@none@1@S@* assess the probability of a given word sequence appearing in text of a language of interest in pattern recognition systems, [[speech recognition]], OCR ([[optical character recognition]]), [[Intelligent Character Recognition]] ([[ICR]]), [[machine translation]] and similar applications@@@@1@36@@danf@17-8-2009
10610670@unknown@formal@none@1@S@* improve retrieval in [[information retrieval]] systems when it is hoped to find similar "documents" (a term for which the conventional meaning is sometimes stretched, depending on the data set) given a single query document and a database of reference documents@@@@1@41@@danf@17-8-2009
10610680@unknown@formal@none@1@S@* improve retrieval performance in genetic sequence analysis as in the [[BLAST]] family of programs@@@@1@15@@danf@17-8-2009
10610690@unknown@formal@none@1@S@* identify the language a text is in or the species a small sequence of DNA was taken from@@@@1@19@@danf@17-8-2009
10610700@unknown@formal@none@1@S@* predict letters or words at random in order to create text, as in the [[dissociated press]] algorithm.@@@@1@18@@danf@17-8-2009
10610710@unknown@formal@none@1@S@== Bias-versus-variance trade-off ==@@@@1@4@@danf@17-8-2009
10610720@unknown@formal@none@1@S@What goes into picking the ''n'' for the ''n''-gram?@@@@1@9@@danf@17-8-2009
10610730@unknown@formal@none@1@S@There are problems of balance weight between ''infrequent grams'' (for example, if a proper name appeared in the training data) and ''frequent grams''.@@@@1@23@@danf@17-8-2009
10610740@unknown@formal@none@1@S@Also, items not seen in the training data will be given a [[probability]] of 0.0 without [[smoothing]].@@@@1@17@@danf@17-8-2009
10610750@unknown@formal@none@1@S@For unseen but plausible data from a sample, one can introduce [[pseudocount]]s.@@@@1@12@@danf@17-8-2009
10610760@unknown@formal@none@1@S@Pseudocounts are generally motivated on Bayesian grounds.@@@@1@7@@danf@17-8-2009
10610770@unknown@formal@none@1@S@=== Smoothing techniques ===@@@@1@4@@danf@17-8-2009
10610780@unknown@formal@none@1@S@* [[Linear interpolation]] (e.g., taking the [[weighted mean]] of the unigram, bigram, and trigram)@@@@1@14@@danf@17-8-2009
10610790@unknown@formal@none@1@S@* [[Good-Turing]] discounting@@@@1@3@@danf@17-8-2009
10610800@unknown@formal@none@1@S@* [[Witten-Bell discounting]]@@@@1@3@@danf@17-8-2009
10610810@unknown@formal@none@1@S@* [[Katz's back-off model]] (trigram)@@@@1@5@@danf@17-8-2009
10610820@unknown@formal@none@1@S@==Google use of N-gram==@@@@1@4@@danf@17-8-2009
10610830@unknown@formal@none@1@S@[[Google]] uses n-gram models for a variety of R&D projects, such as [[statistical machine translation]], [[speech recognition]], [[Spell checker|checking spelling]], [[entity detection]], and [[information extraction|data mining]].@@@@1@26@@danf@17-8-2009
10610840@unknown@formal@none@1@S@In September of 2006 [http://googleresearch.blogspot.com/2006/08/all-our-n-gram-are-belong-to-you.html Google announced] that they made their n-grams [http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2006T13 public] at the [[Linguistic Data Consortium]] ([http://www.ldc.upenn.edu/ LDC]).@@@@1@21@@danf@17-8-2009
10620010@unknown@formal@none@1@S@Noun@@@@1@1@@danf@17-8-2009
10620020@unknown@formal@none@1@S@In [[linguistics]], a '''noun''' is a member of a large, [[open class (linguistics)|open]] [[lexical category]] whose members can occur as the main word in the [[subject (grammar)|subject]] of a [[clause]], the [[object (grammar)|object]] of a [[verb]], or the object of a [[preposition]].@@@@1@42@@danf@17-8-2009
10620030@unknown@formal@none@1@S@Lexical categories are defined in terms of how their members combine with other kinds of expressions.@@@@1@16@@danf@17-8-2009
10620040@unknown@formal@none@1@S@The syntactic rules for nouns differ from language to language.@@@@1@10@@danf@17-8-2009
10620050@unknown@formal@none@1@S@In [[English language|English]], nouns may be defined as those words which can occur with articles and [[adjective|attributive adjectives]] and can function as the [[phrase|head]] of a [[noun phrase]].@@@@1@28@@danf@17-8-2009
10620060@unknown@formal@none@1@S@In [[traditional grammar|traditional]] English grammar, the noun is one of the eight [[parts of speech]].@@@@1@15@@danf@17-8-2009
10620070@unknown@formal@none@1@S@==History==@@@@1@1@@danf@17-8-2009
10620080@unknown@formal@none@1@S@The word comes from the [[Latin]] ''nomen'' meaning "[[name]]".@@@@1@9@@danf@17-8-2009
10620090@unknown@formal@none@1@S@Word classes like nouns were first described by the Sanskrit grammarian [[Panini (grammarian)|{{IAST|Pāṇini}}]] and ancient Greeks like [[Dionysios Thrax]]; and were defined in terms of their [[morphology (linguistics)|morphological]] properties.@@@@1@29@@danf@17-8-2009
10620100@unknown@formal@none@1@S@For example, in Ancient Greek, nouns inflect for [[case (grammar)|grammatical case]], such as dative or accusative.@@@@1@16@@danf@17-8-2009
10620110@unknown@formal@none@1@S@[[Verb]]s, on the other hand, inflect for [[grammatical tense|tenses]], such as past, present or future, while nouns do not.@@@@1@19@@danf@17-8-2009
10620120@unknown@formal@none@1@S@[[Aristotle]] also had a notion of ''onomata'' (nouns) and ''rhemata'' (verbs) which, however, does not exactly correspond with modern notions of nouns and verbs.@@@@1@24@@danf@17-8-2009
10620130@unknown@formal@none@1@S@Vinokurova 2005 has a more detailed discussion of the historical origin of the notion of a noun.@@@@1@17@@danf@17-8-2009
10620140@unknown@formal@none@1@S@==Different definitions of nouns==@@@@1@4@@danf@17-8-2009
10620150@unknown@formal@none@1@S@Expressions of [[natural language]] have properties at different levels.@@@@1@9@@danf@17-8-2009
10620160@unknown@formal@none@1@S@They have ''formal'' properties, like what kinds of [[morphology (linguistics)|morphological]] [[prefix]]es or [[suffix]]es they take and what kinds of other expressions they combine with; but they also have [[semantics|semantic]] properties, i.e. properties pertaining to their meaning.@@@@1@36@@danf@17-8-2009
10620170@unknown@formal@none@1@S@The definition of a noun at the outset of this page is thus a ''formal'', traditional grammatical definition.@@@@1@18@@danf@17-8-2009
10620180@unknown@formal@none@1@S@That definition, for the most part, is considered uncontroversial and furnishes the propensity for certain language users to effectively distinguish most nouns from non-nouns.@@@@1@24@@danf@17-8-2009
10620190@unknown@formal@none@1@S@However, it has the disadvantage that it does not apply to nouns in all languages.@@@@1@15@@danf@17-8-2009
10620200@unknown@formal@none@1@S@For example in [[Russian language|Russian]], there are no definite articles, so one cannot define nouns as words that are modified by definite articles.@@@@1@23@@danf@17-8-2009
10620210@unknown@formal@none@1@S@There are also several attempts of defining nouns in terms of their [[semantics|semantic]] properties.@@@@1@14@@danf@17-8-2009
10620220@unknown@formal@none@1@S@Many of these are controversial, but some are discussed below.@@@@1@10@@danf@17-8-2009
10620230@unknown@formal@none@1@S@===Names for things===@@@@1@3@@danf@17-8-2009
10620240@unknown@formal@none@1@S@In [[Traditional grammar|traditional school grammars]], one often encounters the definition of nouns that they are all and only those expressions that refer to a ''person'', ''place'', ''thing'', ''event'', ''substance'', ''quality'', or ''idea'', etc.@@@@1@33@@danf@17-8-2009
10620250@unknown@formal@none@1@S@This is a ''semantic'' definition.@@@@1@5@@danf@17-8-2009
10620260@unknown@formal@none@1@S@It has been criticized by contemporary linguists as being uninformative.@@@@1@10@@danf@17-8-2009
10620270@unknown@formal@none@1@S@Contemporary linguists generally agree that one cannot successfully define nouns (or other grammatical categories) in terms of what sort of ''object in the world'' they ''[[reference|refer]] to'' or ''[[signification|signify]]''.@@@@1@29@@danf@17-8-2009
10620280@unknown@formal@none@1@S@Part of the [[conundrum]] is that the definition makes use of relatively ''general'' nouns ("thing", "phenomenon", "event") to define what nouns ''are''.@@@@1@22@@danf@17-8-2009
10620290@unknown@formal@none@1@S@The existence of such ''general'' nouns demonstrates that nouns refer to entities that are organized in [[taxonomy|taxonomic]] [[hierarchies]].@@@@1@18@@danf@17-8-2009
10620300@unknown@formal@none@1@S@But other kinds of expressions are also organized into such structured taxonomic relationships.@@@@1@13@@danf@17-8-2009
10620310@unknown@formal@none@1@S@For example the verbs "stroll","saunter", "stride", and "tread" are more specific words than the more ''general'' "walk".@@@@1@17@@danf@17-8-2009
10620320@unknown@formal@none@1@S@Moreover, "walk" is more specific than the verb "move", which, in turn, is less general than "change".@@@@1@17@@danf@17-8-2009
10620330@unknown@formal@none@1@S@But it is unlikely that such taxonomic relationships can be used to ''define'' nouns and verbs.@@@@1@16@@danf@17-8-2009
10620340@unknown@formal@none@1@S@We cannot ''define'' verbs as those words that refer to "changes" or "states", for example, because the nouns ''change'' and ''state'' probably refer to such things, but, of course, aren't verbs.@@@@1@31@@danf@17-8-2009
10620350@unknown@formal@none@1@S@Similarly, nouns like "invasion", "meeting", or "collapse" refer to things that are "done" or "happen".@@@@1@15@@danf@17-8-2009
10620360@unknown@formal@none@1@S@In fact, an influential [[theory]] has it that verbs like "kill" or "die" refer to events, which is among the sort of thing that nouns are supposed to refer to.@@@@1@30@@danf@17-8-2009
10620370@unknown@formal@none@1@S@The point being made here is not that this view of verbs is wrong, but rather that this property of verbs is a poor basis for a ''definition'' of this category, just like the property of ''having wheels'' is a poor basis for a definition of cars (some things that have wheels, such as my suitcase or a jumbo jet, aren't cars).@@@@1@62@@danf@17-8-2009
10620380@unknown@formal@none@1@S@Similarly, adjectives like "yellow" or "difficult" might be thought to refer to qualities, and adverbs like "outside" or "upstairs" seem to refer to places, which are also among the sorts of things nouns can refer to.@@@@1@36@@danf@17-8-2009
10620390@unknown@formal@none@1@S@But verbs, adjectives and adverbs are not nouns, and nouns aren't verbs, adjectives or adverbs.@@@@1@15@@danf@17-8-2009
10620400@unknown@formal@none@1@S@One might argue that "definitions" of this sort really rely on speakers' prior intuitive knowledge of what nouns, verbs and adjectives are, and, so don't really add anything over and beyond this.@@@@1@32@@danf@17-8-2009
10620410@unknown@formal@none@1@S@Speakers' intuitive knowledge of such things might plausibly be based on ''formal'' criteria, such as the traditional grammatical definition of English nouns aforementioned.@@@@1@23@@danf@17-8-2009
10620420@unknown@formal@none@1@S@===Prototypically referential expressions===@@@@1@3@@danf@17-8-2009
10620430@unknown@formal@none@1@S@Another semantic definition of nouns is that they are ''prototypically referential.''@@@@1@11@@danf@17-8-2009
10620440@unknown@formal@none@1@S@That definition is also not very helpful in distinguishing actual nouns from verbs.@@@@1@13@@danf@17-8-2009
10620450@unknown@formal@none@1@S@But it may still correctly identify a core property of nounhood.@@@@1@11@@danf@17-8-2009
10620460@unknown@formal@none@1@S@For example, we will tend to use nouns like "fool" and "car" when we wish to refer to fools and cars, respectively.@@@@1@22@@danf@17-8-2009
10620470@unknown@formal@none@1@S@The notion that this is '''prototypical''' reflects the fact that such nouns can be used, even though nothing with the corresponding property is referred to:@@@@1@25@@danf@17-8-2009
10620480@unknown@formal@none@1@S@:John is no '''fool'''.@@@@1@4@@danf@17-8-2009
10620490@unknown@formal@none@1@S@:If I had a '''car''', I'd go to Marrakech.@@@@1@9@@danf@17-8-2009
10620500@unknown@formal@none@1@S@The first sentence above doesn't refer to any fools, nor does the second one refer to any particular car.@@@@1@19@@danf@17-8-2009
10620510@unknown@formal@none@1@S@===Predicates with identity criteria===@@@@1@4@@danf@17-8-2009
10620520@unknown@formal@none@1@S@The British logician [[Peter Thomas Geach]] proposed a very subtle semantic definition of nouns.@@@@1@14@@danf@17-8-2009
10620530@unknown@formal@none@1@S@He noticed that adjectives like "same" can modify nouns, but no other kinds of parts of speech, like [[verbs]] or [[adjectives]].@@@@1@21@@danf@17-8-2009
10620540@unknown@formal@none@1@S@Not only that, but there also doesn't seem to be any ''other'' expressions with similar meaning that can modify verbs and adjectives.@@@@1@22@@danf@17-8-2009
10620550@unknown@formal@none@1@S@Consider the following examples.@@@@1@4@@danf@17-8-2009
10620560@unknown@formal@none@1@S@: Good: John and Bill participated in the '''same''' fight.@@@@1@10@@danf@17-8-2009
10620570@unknown@formal@none@1@S@: Bad:@@@@1@2@@danf@17-8-2009
10620580@unknown@formal@none@1@S@*John and Bill '''samely''' fought.@@@@1@5@@danf@17-8-2009
10620590@unknown@formal@none@1@S@There is no English adverb "samely".@@@@1@6@@danf@17-8-2009
10620600@unknown@formal@none@1@S@In some other languages, like Czech, however there are adverbs corresponding to "samely".@@@@1@13@@danf@17-8-2009
10620610@unknown@formal@none@1@S@Hence, in Czech, the translation of the last sentence would be fine; however, it would mean that John and Bill fought ''in the same way'': not that they participated in the ''same fight''.@@@@1@33@@danf@17-8-2009
10620620@unknown@formal@none@1@S@Geach proposed that we could explain this, if nouns denote logical [[predicate (grammar)|predicate]]s with '''identity criteria'''.@@@@1@16@@danf@17-8-2009
10620630@unknown@formal@none@1@S@An identity criterion would allow us to conclude, for example, that "person x at time 1 is ''the same person'' as person y at time 2".@@@@1@26@@danf@17-8-2009
10620640@unknown@formal@none@1@S@Different nouns can have different identity criteria.@@@@1@7@@danf@17-8-2009
10620650@unknown@formal@none@1@S@A well known example of this is due to Gupta:@@@@1@10@@danf@17-8-2009
10620660@unknown@formal@none@1@S@:National Airlines transported 2 million '''passengers''' in 1979.@@@@1@8@@danf@17-8-2009
10620670@unknown@formal@none@1@S@:National Airlines transported (at least) 2 million '''persons''' in 1979.@@@@1@10@@danf@17-8-2009
10620680@unknown@formal@none@1@S@Given that, in general, all passengers are persons, the last sentence above ought to follow logically from the first one.@@@@1@20@@danf@17-8-2009
10620690@unknown@formal@none@1@S@But it doesn't.@@@@1@3@@danf@17-8-2009
10620700@unknown@formal@none@1@S@It is easy to imagine, for example, that on average, every person who travelled with National Airlines in 1979, travelled with them twice.@@@@1@23@@danf@17-8-2009
10620710@unknown@formal@none@1@S@In that case, one would say that the airline transported 2 million ''passengers'' but only 1 million ''persons''.@@@@1@18@@danf@17-8-2009
10620720@unknown@formal@none@1@S@Thus, the way that we count ''passengers'' isn't necessarily the same as the way that we count ''persons''.@@@@1@18@@danf@17-8-2009
10620730@unknown@formal@none@1@S@Put somewhat differently: At two different times, ''you'' may correspond to two distinct ''passengers'', even though you are one and the same person.@@@@1@23@@danf@17-8-2009
10620740@unknown@formal@none@1@S@For a precise definition of ''identity criteria'', see Gupta.@@@@1@9@@danf@17-8-2009
10620750@unknown@formal@none@1@S@Recently, Baker has proposed that Geach's definition of nouns in terms of identity criteria allows us to ''explain'' the characteristic properties of nouns.@@@@1@23@@danf@17-8-2009
10620760@unknown@formal@none@1@S@He argues that nouns can co-occur with (in-)definite articles and numerals, and are "prototypically referential" ''because'' they are all and only those [[parts of speech]] that provide identity criteria.@@@@1@29@@danf@17-8-2009
10620770@unknown@formal@none@1@S@Baker's proposals are quite new, and linguists are still evaluating them.@@@@1@11@@danf@17-8-2009
10620780@unknown@formal@none@1@S@==Classification of nouns in English==@@@@1@5@@danf@17-8-2009
10620790@unknown@formal@none@1@S@===Proper nouns and common nouns===@@@@1@5@@danf@17-8-2009
10620800@unknown@formal@none@1@S@''Proper nouns'' (also called ''[[proper name]]s'') are nouns representing unique entities (such as ''London'', ''Universe'' or ''John''), as distinguished from common nouns which describe a class of entities (such as ''city'', ''planet'' or ''person'').@@@@1@34@@danf@17-8-2009
10620810@unknown@formal@none@1@S@In [[English language|English]] and most other languages that use the [[Latin alphabet]], proper nouns are usually [[capitalization|capitalized]].@@@@1@17@@danf@17-8-2009
10620820@unknown@formal@none@1@S@Languages differ in whether most elements of multiword proper nouns are capitalised (e.g., American English ''House of Representatives'') or only the initial element (e.g., Slovenian ''Državni zbor'' 'National Assembly').@@@@1@29@@danf@17-8-2009
10620830@unknown@formal@none@1@S@In [[German language|German]], nouns of all types are capitalized.@@@@1@9@@danf@17-8-2009
10620840@unknown@formal@none@1@S@The convention of capitalizing ''all'' nouns was previously used in English, but ended circa 1800.@@@@1@15@@danf@17-8-2009
10620850@unknown@formal@none@1@S@In America, the shift in capitalization is recorded in several noteworthy documents.@@@@1@12@@danf@17-8-2009
10620860@unknown@formal@none@1@S@The end (but not the beginning) of the [[United States Declaration of Independence#Annotated text of the Declaration|Declaration of Independence]] (1776) and all of the [[United States Constitution|Constitution]] (1787) show nearly all nouns capitalized, the [[United States Bill of Rights#Text of the Bill of Rights|Bill of Rights]] (1789) capitalizes a few common nouns but not most of them, and the [[Thirteenth Amendment to the United States Constitution|Thirteenth Constitutional Amendment]] (1865) only capitalizes proper nouns.@@@@1@73@@danf@17-8-2009
10620870@unknown@formal@none@1@S@Sometimes the same word can function as both a common noun and a proper noun, where one such entity is special.@@@@1@21@@danf@17-8-2009
10620880@unknown@formal@none@1@S@For example the common noun ''god'' denotes all deities, while the proper noun ''God'' references the [[monotheism|monotheistic]] [[God]] specifically.@@@@1@19@@danf@17-8-2009
10620890@unknown@formal@none@1@S@Owing to the essentially arbitrary nature of [[Orthography|orthographic]] classification and the existence of variant authorities and adopted [[Style guide|''house styles'']], questionable capitalization of words is not uncommon, even in respected newspapers and magazines.@@@@1@33@@danf@17-8-2009
10620900@unknown@formal@none@1@S@Most publishers, however, properly require ''consistency'', at least within the same document, in applying their specified standard.@@@@1@17@@danf@17-8-2009
10620910@unknown@formal@none@1@S@The common meaning of the word or words constituting a proper noun may be unrelated to the object to which the proper noun refers.@@@@1@24@@danf@17-8-2009
10620920@unknown@formal@none@1@S@For example, someone might be named "Tiger Smith" despite being neither a [[tiger]] nor a [[smith (metalwork)|smith]].@@@@1@17@@danf@17-8-2009
10620930@unknown@formal@none@1@S@For this reason, proper nouns are usually not [[translation|translated]] between languages, although they may be [[transliteration|transliterated]].@@@@1@16@@danf@17-8-2009
10620940@unknown@formal@none@1@S@For example, the German surname ''Knödel'' becomes ''Knodel'' or ''Knoedel'' in English (not the literal ''Dumpling'').@@@@1@16@@danf@17-8-2009
10620950@unknown@formal@none@1@S@However, the [[Transliteration|transcription]] of place names and the names of [[monarch]]s, [[pope]]s, and non-contemporary [[author]]s is common and sometimes universal.@@@@1@20@@danf@17-8-2009
10620960@unknown@formal@none@1@S@For instance, the [[Portuguese language|Portuguese]] word ''Lisboa'' becomes ''[[Lisbon]]'' in [[English language|English]]; the English ''London'' becomes ''Londres'' in French; and the [[ancient Greek|Greek]] ''Aristotelēs'' becomes [[Aristotle]] in English.@@@@1@28@@danf@17-8-2009
10620970@unknown@formal@none@1@S@===Countable and uncountable nouns===@@@@1@4@@danf@17-8-2009
10620980@unknown@formal@none@1@S@''Count nouns'' are common nouns that can take a [[plural]], can combine with [[numerals]] or [[quantifiers]] (e.g. "one", "two", "several", "every", "most"), and can take an indefinite article ("a" or "an").@@@@1@31@@danf@17-8-2009
10620990@unknown@formal@none@1@S@Examples of count nouns are "chair", "nose", and "occasion".@@@@1@9@@danf@17-8-2009
10621000@unknown@formal@none@1@S@''Mass nouns'' (or ''non-count nouns'') differ from count nouns in precisely that respect: they can't take plural or combine with number words or quantifiers.@@@@1@24@@danf@17-8-2009
10621010@unknown@formal@none@1@S@Examples from English include "laughter", "cutlery", "helium", and "furniture".@@@@1@9@@danf@17-8-2009
10621020@unknown@formal@none@1@S@For example, it is not possible to refer to "a furniture" or "three furnitures".@@@@1@14@@danf@17-8-2009
10621030@unknown@formal@none@1@S@This is true even though the pieces of furniture comprising "furniture" could be counted.@@@@1@14@@danf@17-8-2009
10621040@unknown@formal@none@1@S@Thus the distinction between mass and count nouns shouldn't be made in terms of what sorts of things the nouns ''refer'' to, but rather in terms of how the nouns ''present'' these entities.@@@@1@33@@danf@17-8-2009
10621050@unknown@formal@none@1@S@===Collective nouns===@@@@1@2@@danf@17-8-2009
10621060@unknown@formal@none@1@S@''Collective nouns'' are nouns that refer to ''groups'' consisting of more than one individual or entity, even when they are inflected for the [[Grammatical number|singular]].@@@@1@25@@danf@17-8-2009
10621070@unknown@formal@none@1@S@Examples include "committee", "herd", and "school" (of herring).@@@@1@8@@danf@17-8-2009
10621080@unknown@formal@none@1@S@These nouns have slightly different grammatical properties than other nouns.@@@@1@10@@danf@17-8-2009
10621090@unknown@formal@none@1@S@For example, the [[noun phrases]] that they [[head (syntax)|head]] can serve as the [[subject (grammar)|subject]] of a [[collective predicate]], even when they are inflected for the singular.@@@@1@27@@danf@17-8-2009
10621100@unknown@formal@none@1@S@A [[collective predicate]] is a predicate that normally can't take a singular subject.@@@@1@13@@danf@17-8-2009
10621110@unknown@formal@none@1@S@An example of the latter is "talked to each other".@@@@1@10@@danf@17-8-2009
10621120@unknown@formal@none@1@S@:Good: The '''boys''' talked to each other.@@@@1@7@@danf@17-8-2009
10621130@unknown@formal@none@1@S@:Bad: *The '''boy''' talked to each other.@@@@1@7@@danf@17-8-2009
10621140@unknown@formal@none@1@S@:Good: The '''committee''' talked to each other.@@@@1@7@@danf@17-8-2009
10621150@unknown@formal@none@1@S@===Concrete nouns and abstract nouns===@@@@1@5@@danf@17-8-2009
10621160@unknown@formal@none@1@S@''Concrete nouns'' refer to [[physical bodies]] which you use at least one of your [[sense]]s to observe.@@@@1@17@@danf@17-8-2009
10621170@unknown@formal@none@1@S@For instance, "chair", "apple", or "Janet".@@@@1@6@@danf@17-8-2009
10621180@unknown@formal@none@1@S@''Abstract nouns'' on the other hand refer to [[abstract object]]s, that is ideas or concepts, such as "justice" or "hate".@@@@1@20@@danf@17-8-2009
10621190@unknown@formal@none@1@S@While this distinction is sometimes useful, the boundary between the two of them is not always clear; consider, for example, the noun "art".@@@@1@23@@danf@17-8-2009
10621200@unknown@formal@none@1@S@In English, many abstract nouns are formed by adding noun-forming suffixes ("-ness", "-ity", "-tion") to adjectives or verbs.@@@@1@18@@danf@17-8-2009
10621210@unknown@formal@none@1@S@Examples are "happiness", "circulation" and "serenity".@@@@1@6@@danf@17-8-2009
10621220@unknown@formal@none@1@S@==Nouns and pronouns==@@@@1@3@@danf@17-8-2009
10621230@unknown@formal@none@1@S@[[Noun phrase]]s can typically be replaced by [[pronoun]]s, such as "he", "it", "which", and "those", in order to avoid repetition or explicit identification, or for other reasons.@@@@1@27@@danf@17-8-2009
10621240@unknown@formal@none@1@S@For example, in the sentence "Janet thought that he was weird", the word "he" is a pronoun standing in place of the name of the person in question.@@@@1@28@@danf@17-8-2009
10621250@unknown@formal@none@1@S@The English word ''one'' can replace parts of [[noun phrase]]s, and it sometimes stands in for a noun.@@@@1@18@@danf@17-8-2009
10621260@unknown@formal@none@1@S@An example is given below:@@@@1@5@@danf@17-8-2009
10621270@unknown@formal@none@1@S@: John's car is newer than ''the one'' that Bill has.@@@@1@11@@danf@17-8-2009
10621280@unknown@formal@none@1@S@But ''one'' can also stand in for bigger subparts of a noun phrase.@@@@1@13@@danf@17-8-2009
10621290@unknown@formal@none@1@S@For example, in the following example, ''one'' can stand in for ''new car''.@@@@1@13@@danf@17-8-2009
10621300@unknown@formal@none@1@S@: This new car is cheaper than ''that one''.@@@@1@9@@danf@17-8-2009
10621310@unknown@formal@none@1@S@==Substantive as a word for "noun"==@@@@1@6@@danf@17-8-2009
10621320@unknown@formal@none@1@S@Starting with old [[Latin language|Latin]] grammars, many European languages use some form of the word ''substantive'' as the basic term for noun.@@@@1@22@@danf@17-8-2009
10621330@unknown@formal@none@1@S@Nouns in the dictionaries of such languages are demarked by the abbreviation "s" instead of "n", which may be used for proper nouns instead.@@@@1@24@@danf@17-8-2009
10621340@unknown@formal@none@1@S@This corresponds to those grammars in which nouns and adjectives phase into each other in more areas than, for example, the English term [[Predicative_adjective#Predicative_adjective|predicate adjective]] entails.@@@@1@26@@danf@17-8-2009
10621350@unknown@formal@none@1@S@In French and Spanish, for example, adjectives frequently act as nouns referring to people who have the characteristics of the adjective.@@@@1@21@@danf@17-8-2009
10621360@unknown@formal@none@1@S@An example in English is:@@@@1@5@@danf@17-8-2009
10621370@unknown@formal@none@1@S@: The ''poor'' you have always with you.@@@@1@8@@danf@17-8-2009
10621380@unknown@formal@none@1@S@Similarly, an adjective can also be used for a whole group or organization of people:@@@@1@15@@danf@17-8-2009
10621390@unknown@formal@none@1@S@: The Socialist ''International''.@@@@1@4@@danf@17-8-2009
10621400@unknown@formal@none@1@S@Hence, these words are substantives that are usually adjectives in English.@@@@1@11@@danf@17-8-2009
10630010@unknown@formal@none@1@S@Ontology (information science)@@@@1@3@@danf@17-8-2009
10630020@unknown@formal@none@1@S@In both [[computer science]] and [[information science]], an '''ontology''' is a formal representation of a set of concepts within a [[Domain of discourse|domain]] and the relationships between those concepts.@@@@1@29@@danf@17-8-2009
10630030@unknown@formal@none@1@S@It is used to [[Reasoning|reason]] about the properties of that domain, and may be used to define the domain.@@@@1@19@@danf@17-8-2009
10630040@unknown@formal@none@1@S@Ontologies are used in [[artificial intelligence]], the [[Semantic Web]], [[software engineering]], [[biomedical informatics]], [[library science]], and [[information architecture]] as a form of [[knowledge representation]] about the world or some part of it.@@@@1@32@@danf@17-8-2009
10630050@unknown@formal@none@1@S@Common components of ontologies include:@@@@1@5@@danf@17-8-2009
10630060@unknown@formal@none@1@S@* Individuals: instances or objects (the basic or "ground level" objects)@@@@1@11@@danf@17-8-2009
10630070@unknown@formal@none@1@S@* [[Class]]es: [[set (computer science)|set]]s, collections, concepts or types of objects@@@@1@11@@danf@17-8-2009
10630080@unknown@formal@none@1@S@* [[Attribute (computing)|Attribute]]s: properties, features, characteristics, or parameters that objects (and classes) can have@@@@1@14@@danf@17-8-2009
10630090@unknown@formal@none@1@S@* [[Relation (mathematics)|Relations]]: ways that classes and objects can be related to one another@@@@1@14@@danf@17-8-2009
10630100@unknown@formal@none@1@S@* Function terms: complex structures formed from certain relations that can be used in place of an individual term in a statement@@@@1@22@@danf@17-8-2009
10630110@unknown@formal@none@1@S@* Restrictions: formally stated descriptions of what must be true in order for some assertion to be accepted as input@@@@1@20@@danf@17-8-2009
10630120@unknown@formal@none@1@S@* Rules: statements in the form of an if-then (antecedent-consequent) sentence that describe the logical inferences that can be drawn from an assertion in a particular form@@@@1@27@@danf@17-8-2009
10630130@unknown@formal@none@1@S@* Axioms: assertions (including rules) in a logical form that together comprise the overall theory that the ontology describes in its domain of application.@@@@1@24@@danf@17-8-2009
10630140@unknown@formal@none@1@S@This definition differs from that of "axioms" in generative grammar and formal logic.@@@@1@13@@danf@17-8-2009
10630150@unknown@formal@none@1@S@In these disciplines, axioms include only statements asserted as ''a priori'' knowledge.@@@@1@12@@danf@17-8-2009
10630160@unknown@formal@none@1@S@As used here, "axioms" also include the theory derived from axiomatic statements.@@@@1@12@@danf@17-8-2009
10630170@unknown@formal@none@1@S@* [[Event (philosophy)|Events]]: the changing of attributes or relations@@@@1@9@@danf@17-8-2009
10630180@unknown@formal@none@1@S@Ontologies are commonly encoded using [[ontology language]]s.@@@@1@7@@danf@17-8-2009
10630190@unknown@formal@none@1@S@== Elements ==@@@@1@3@@danf@17-8-2009
10630200@unknown@formal@none@1@S@Contemporary ontologies share many structural similarities, regardless of the language in which they are expressed.@@@@1@15@@danf@17-8-2009
10630210@unknown@formal@none@1@S@As mentioned above, most ontologies describe individuals (instances), classes (concepts), attributes, and relations.@@@@1@13@@danf@17-8-2009
10630220@unknown@formal@none@1@S@In this section each of these components is discussed in turn.@@@@1@11@@danf@17-8-2009
10630230@unknown@formal@none@1@S@=== Individuals ===@@@@1@3@@danf@17-8-2009
10630240@unknown@formal@none@1@S@Individuals (instances) are the basic, "ground level" components of an ontology.@@@@1@11@@danf@17-8-2009
10630250@unknown@formal@none@1@S@The individuals in an ontology may include concrete objects such as people, animals, tables, automobiles, molecules, and planets, as well as abstract individuals such as numbers and words.@@@@1@28@@danf@17-8-2009
10630260@unknown@formal@none@1@S@Strictly speaking, an ontology need not include any individuals, but one of the general purposes of an ontology is to provide a means of classifying individuals, even if those individuals are not explicitly part of the ontology.@@@@1@37@@danf@17-8-2009
10630270@unknown@formal@none@1@S@In formal extensional ontologies, only the utterances of words and numbers are considered individuals – the numbers and names themselves are classes.@@@@1@22@@danf@17-8-2009
10630280@unknown@formal@none@1@S@In a 4D ontology, an individual is identified by its spatio-temporal extent.@@@@1@12@@danf@17-8-2009
10630290@unknown@formal@none@1@S@Examples of formal extensional ontologies are [[ISO 15926]] and the model in development by the [[IDEAS Group]].@@@@1@17@@danf@17-8-2009
10630300@unknown@formal@none@1@S@=== Classes ===@@@@1@3@@danf@17-8-2009
10630310@unknown@formal@none@1@S@Classes – concepts that are also called ''type'', ''sort'', ''category'', and ''kind'' – are abstract groups, sets, or collections of objects.@@@@1@21@@danf@17-8-2009
10630320@unknown@formal@none@1@S@They may contain individuals, other classes, or a combination of both.@@@@1@11@@danf@17-8-2009
10630330@unknown@formal@none@1@S@Some examples of classes:@@@@1@4@@danf@17-8-2009
10630340@unknown@formal@none@1@S@* ''Person'', the class of all people@@@@1@7@@danf@17-8-2009
10630350@unknown@formal@none@1@S@* ''Vehicle'', the class of all vehicles@@@@1@7@@danf@17-8-2009
10630360@unknown@formal@none@1@S@* ''Car'', the class of all cars@@@@1@7@@danf@17-8-2009
10630370@unknown@formal@none@1@S@* ''Class'', representing the class of all classes@@@@1@8@@danf@17-8-2009
10630380@unknown@formal@none@1@S@* ''Thing'', representing the class of all things@@@@1@8@@danf@17-8-2009
10630390@unknown@formal@none@1@S@Ontologies vary on whether classes can contain other classes, whether a class can belong to itself, whether there is a universal class (that is, a class containing everything), etc.@@@@1@29@@danf@17-8-2009
10630400@unknown@formal@none@1@S@Sometimes restrictions along these lines are made in order to avoid certain well-known [[paradox]]es.@@@@1@14@@danf@17-8-2009
10630410@unknown@formal@none@1@S@The classes of an ontology may be [[extensional]] or [[intensional]] in nature.@@@@1@12@@danf@17-8-2009
10630420@unknown@formal@none@1@S@A class is extensional if and only if it is characterized solely by its membership.@@@@1@15@@danf@17-8-2009
10630430@unknown@formal@none@1@S@More precisely, a class C is extensional if and only if for any class C', if C' has exactly the same members as C, then C and C' are identical.@@@@1@30@@danf@17-8-2009
10630440@unknown@formal@none@1@S@If a class does not satisfy this condition, then it is intensional.@@@@1@12@@danf@17-8-2009
10630450@unknown@formal@none@1@S@While extensional classes are more well-behaved and well-understood mathematically, as well as less problematic philosophically, they do not permit the fine grained distinctions that ontologies often need to make.@@@@1@29@@danf@17-8-2009
10630460@unknown@formal@none@1@S@For example, an ontology may want to distinguish between the class of all creatures with a kidney and the class of all creatures with a heart, even if these classes happen to have exactly the same members.@@@@1@37@@danf@17-8-2009
10630470@unknown@formal@none@1@S@In the upper ontologies mentioned above, the classes are defined intensionally.@@@@1@11@@danf@17-8-2009
10630480@unknown@formal@none@1@S@Intensionally defined classes usually have necessary conditions associated with membership in each class.@@@@1@13@@danf@17-8-2009
10630490@unknown@formal@none@1@S@Some classes may also have sufficient conditions, and in those cases the combination of necessary and sufficient conditions make that class a fully ''defined'' class.@@@@1@25@@danf@17-8-2009
10630500@unknown@formal@none@1@S@Importantly, a class can subsume or be subsumed by other classes; a class subsumed by another is called a ''subclass'' of the subsuming class.@@@@1@24@@danf@17-8-2009
10630510@unknown@formal@none@1@S@For example, ''Vehicle'' subsumes ''Car'', since (necessarily) anything that is a member of the latter class is a member of the former.@@@@1@22@@danf@17-8-2009
10630520@unknown@formal@none@1@S@The subsumption relation is used to create a hierarchy of classes, typically with a maximally general class like ''Thing'' at the top, and very specific classes like ''2002 Ford Explorer'' at the bottom.@@@@1@33@@danf@17-8-2009
10630530@unknown@formal@none@1@S@The critically important consequence of the subsumption relation is the inheritance of properties from the parent (subsuming) class to the child (subsumed) class.@@@@1@23@@danf@17-8-2009
10630540@unknown@formal@none@1@S@Thus, anything that is necessarily true of a parent class is also necessarily true of all of its subsumed child classes.@@@@1@21@@danf@17-8-2009
10630550@unknown@formal@none@1@S@In some ontologies, a class is only allowed to have one parent (''single inheritance''), but in most ontologies, classes are allowed to have any number of parents (''multiple inheritance''), and in the latter case all necessary properties of each parent are inherited by the subsumed child class.@@@@1@47@@danf@17-8-2009
10630560@unknown@formal@none@1@S@Thus a particular class of animal (''HouseCat'') may be a child of the class ''Cat'' and also a child of the class ''Pet''.@@@@1@23@@danf@17-8-2009
10630570@unknown@formal@none@1@S@A partition is a set of related classes and associated rules that allow objects to be placed into the appropriate class.@@@@1@21@@danf@17-8-2009
10630580@unknown@formal@none@1@S@For example, to the right is the partial diagram of an ontology that has a partition of the ''Car'' class into the classes ''2-Wheel Drive'' and ''4-Wheel Drive''.@@@@1@28@@danf@17-8-2009
10630590@unknown@formal@none@1@S@The partition rule determines if a particular car is placed in the ''2-Wheel Drive'' or the ''4-Wheel Drive'' class.@@@@1@19@@danf@17-8-2009
10630600@unknown@formal@none@1@S@If the partition rule(s) guarantee that a single ''Car'' cannot be in both classes, then the partition is called a disjoint partition.@@@@1@22@@danf@17-8-2009
10630610@unknown@formal@none@1@S@If the partition rules ensure that every concrete object in the super-class is an instance of at least one of the partition classes, then the partition is called an exhaustive partition.@@@@1@31@@danf@17-8-2009
10630620@unknown@formal@none@1@S@=== Attributes ===@@@@1@3@@danf@17-8-2009
10630630@unknown@formal@none@1@S@Objects in the ontology can be described by assigning attributes to them.@@@@1@12@@danf@17-8-2009
10630640@unknown@formal@none@1@S@Each attribute has at least a name and a value, and is used to store information that is specific to the object it is attached to.@@@@1@26@@danf@17-8-2009
10630650@unknown@formal@none@1@S@For example the Ford Explorer object has attributes such as:@@@@1@10@@danf@17-8-2009
10630660@unknown@formal@none@1@S@* ''Name'': Ford Explorer@@@@1@4@@danf@17-8-2009
10630670@unknown@formal@none@1@S@* ''Number-of-doors'': 4@@@@1@3@@danf@17-8-2009
10630680@unknown@formal@none@1@S@* ''Engine'': {4.0L, 4.6L}@@@@1@4@@danf@17-8-2009
10630690@unknown@formal@none@1@S@* ''Transmission'': 6-speed@@@@1@3@@danf@17-8-2009
10630700@unknown@formal@none@1@S@The value of an attribute can be a complex [[data type]]; in this example, the value of the attribute called ''Engine'' is a list of values, not just a single value.@@@@1@31@@danf@17-8-2009
10630710@unknown@formal@none@1@S@If you did not define attributes for the concepts you would have either a [[taxonomy]] (if [[hyponym]] relationships exist between concepts) or a '''controlled vocabulary'''.@@@@1@25@@danf@17-8-2009
10630720@unknown@formal@none@1@S@These are useful, but are not considered true ontologies.@@@@1@9@@danf@17-8-2009
10630730@unknown@formal@none@1@S@===Relationships===@@@@1@1@@danf@17-8-2009
10630740@unknown@formal@none@1@S@An important use of attributes is to describe the relationships (also known as relations) between objects in the ontology.@@@@1@19@@danf@17-8-2009
10630750@unknown@formal@none@1@S@Typically a relation is an attribute whose value is another object in the ontology.@@@@1@14@@danf@17-8-2009
10630760@unknown@formal@none@1@S@For example in the ontology that contains the Ford Explorer and the [[Ford Bronco]], the Ford Bronco object might have the following attribute:@@@@1@23@@danf@17-8-2009
10630770@unknown@formal@none@1@S@* ''Successor'': Ford Explorer@@@@1@4@@danf@17-8-2009
10630780@unknown@formal@none@1@S@This tells us that the Explorer is the model that replaced the Bronco.@@@@1@13@@danf@17-8-2009
10630790@unknown@formal@none@1@S@Much of the power of ontologies comes from the ability to describe these relations.@@@@1@14@@danf@17-8-2009
10630800@unknown@formal@none@1@S@Together, the set of relations describes the [[semantics]] of the domain.@@@@1@11@@danf@17-8-2009
10630810@unknown@formal@none@1@S@The most important type of relation is the [[subsumption]] relation (''is-[[superclass]]-of'', the converse of ''[[is-a]]'', ''is-subtype-of'' or ''is-[[subclass]]-of'').@@@@1@18@@danf@17-8-2009
10630820@unknown@formal@none@1@S@This defines which objects are members of classes of objects.@@@@1@10@@danf@17-8-2009
10630830@unknown@formal@none@1@S@For example we have already seen that the Ford Explorer ''is-a'' 4-wheel drive, which in turn ''is-a'' Car:@@@@1@18@@danf@17-8-2009
10630840@unknown@formal@none@1@S@The addition of the is-a relationships has created a hierarchical [[taxonomy]]; a tree-like structure (or, more generally, a [[partially ordered set]]) that clearly depicts how objects relate to one another.@@@@1@30@@danf@17-8-2009
10630850@unknown@formal@none@1@S@In such a structure, each object is the 'child' of a 'parent class' (Some languages restrict the is-a relationship to one parent for all nodes, but many do not).@@@@1@29@@danf@17-8-2009
10630860@unknown@formal@none@1@S@Another common type of relations is the [[meronymy]] relation, written as ''part-of'', that represents how objects combine together to form composite objects.@@@@1@22@@danf@17-8-2009
10630870@unknown@formal@none@1@S@For example, if we extended our example ontology to include objects like Steering Wheel, we would say that "Steering Wheel is-part-of Ford Explorer" since a steering wheel is one of the components of a Ford Explorer.@@@@1@36@@danf@17-8-2009
10630880@unknown@formal@none@1@S@If we introduce meronymy relationships to our ontology, we find that this simple and elegant tree structure quickly becomes complex and significantly more difficult to interpret manually.@@@@1@27@@danf@17-8-2009
10630890@unknown@formal@none@1@S@It is not difficult to understand why; an entity that is described as 'part of' another entity might also be 'part of' a third entity.@@@@1@25@@danf@17-8-2009
10630900@unknown@formal@none@1@S@Consequently, entities may have more than one parent.@@@@1@8@@danf@17-8-2009
10630910@unknown@formal@none@1@S@The structure that emerges is known as a [[directed acyclic graph]] (DAG).@@@@1@12@@danf@17-8-2009
10630920@unknown@formal@none@1@S@As well as the standard is-a and part-of relations, ontologies often include additional types of relation that further refine the semantics they model.@@@@1@23@@danf@17-8-2009
10630930@unknown@formal@none@1@S@These relations are often domain-specific and are used to answer particular types of question.@@@@1@14@@danf@17-8-2009
10630940@unknown@formal@none@1@S@For example in the domain of automobiles, we might define a ''made-in'' relationship which tells us where each car is built.@@@@1@21@@danf@17-8-2009
10630950@unknown@formal@none@1@S@So the Ford Explorer is ''made-in'' [[Louisville, Kentucky|Louisville]].@@@@1@8@@danf@17-8-2009
10630960@unknown@formal@none@1@S@The ontology may also know that Louisville is-in [[Kentucky]] and Kentucky is-a state of the [[United States|USA]].@@@@1@17@@danf@17-8-2009
10630970@unknown@formal@none@1@S@Software using this ontology could now answer a question like "which cars are made in the U.S.?"@@@@1@17@@danf@17-8-2009
10630980@unknown@formal@none@1@S@== Domain ontologies and upper ontologies ==@@@@1@7@@danf@17-8-2009
10630990@unknown@formal@none@1@S@A domain ontology (or domain-specific ontology) models a specific domain, or part of the world.@@@@1@15@@danf@17-8-2009
10631000@unknown@formal@none@1@S@It represents the particular meanings of terms as they apply to that domain.@@@@1@13@@danf@17-8-2009
10631010@unknown@formal@none@1@S@For example the word ''[[card]]'' has many different meanings.@@@@1@9@@danf@17-8-2009
10631020@unknown@formal@none@1@S@An ontology about the domain of [[poker]] would model the "[[playing card]]" meaning of the word, while an ontology about the domain of [[computer hardware]] would model the "[[punch card]]" and "[[video card]]" meanings.@@@@1@34@@danf@17-8-2009
10631030@unknown@formal@none@1@S@An [[Upper ontology (computer science)|upper ontology]] (or foundation ontology) is a model of the common objects that are generally applicable across a wide range of domain ontologies.@@@@1@27@@danf@17-8-2009
10631040@unknown@formal@none@1@S@It contains a [[core glossary]] in whose terms objects in a set of domains can be described.@@@@1@17@@danf@17-8-2009
10631050@unknown@formal@none@1@S@There are several standardized upper ontologies available for use, including [[Dublin Core]], [[General Formal Ontology|GFO]], [[OpenCyc]]/[[ResearchCyc]], [[Suggested Upper Merged Ontology|SUMO]], and [http://www.loa-cnr.it/DOLCE.html DOLCE]l.@@@@1@23@@danf@17-8-2009
10631060@unknown@formal@none@1@S@[[WordNet]], while considered an upper ontology by some, is not an ontology: it is a unique combination of a [[taxonomy]] and a controlled vocabulary (see above, under Attributes).@@@@1@28@@danf@17-8-2009
10631070@unknown@formal@none@1@S@The [[Gellish]] ontology is an example of a combination of an upper and a domain ontology.@@@@1@16@@danf@17-8-2009
10631080@unknown@formal@none@1@S@Since domain ontologies represent concepts in very specific and often eclectic ways, they are often incompatible.@@@@1@16@@danf@17-8-2009
10631090@unknown@formal@none@1@S@As systems that rely on domain ontologies expand, they often need to merge domain ontologies into a more general representation.@@@@1@20@@danf@17-8-2009
10631100@unknown@formal@none@1@S@This presents a challenge to the ontology designer.@@@@1@8@@danf@17-8-2009
10631110@unknown@formal@none@1@S@Different ontologies in the same domain can also arise due to different perceptions of the domain based on cultural background, education, ideology, or because a different representation language was chosen.@@@@1@30@@danf@17-8-2009
10631120@unknown@formal@none@1@S@At present, merging ontologies is a largely manual process and therefore time-consuming and expensive.@@@@1@14@@danf@17-8-2009
10631130@unknown@formal@none@1@S@Using a foundation ontology to provide a common definition of core terms can make this process manageable.@@@@1@17@@danf@17-8-2009
10631140@unknown@formal@none@1@S@There are studies on generalized techniques for merging ontologies, but this area of research is still largely theoretical.@@@@1@18@@danf@17-8-2009
10631150@unknown@formal@none@1@S@== Ontology languages ==@@@@1@4@@danf@17-8-2009
10631160@unknown@formal@none@1@S@An [[ontology language]] is a [[formal language]] used to encode the ontology.@@@@1@12@@danf@17-8-2009
10631170@unknown@formal@none@1@S@There are a number of such languages for ontologies, both proprietary and standards-based:@@@@1@13@@danf@17-8-2009
10631180@unknown@formal@none@1@S@* [[Web Ontology Language|OWL]] is a language for making ontological statements, developed as a follow-on from [[Resource Description Framework|RDF]] and [[RDFS]], as well as earlier ontology language projects including [[Ontology Inference Layer|OIL]], [[DARPA Agent Markup Language|DAML]] and [[DAMLplusOIL|DAML+OIL]].@@@@1@38@@danf@17-8-2009
10631190@unknown@formal@none@1@S@OWL is intended to be used over the [[World Wide Web]], and all its elements (classes, properties and individuals) are defined as RDF [[resource (Web)|resources]], and identified by [[Uniform Resource Identifier|URI]]s.@@@@1@31@@danf@17-8-2009
10631200@unknown@formal@none@1@S@* [[KIF]] is a syntax for [[first-order logic]] that is based on [[S-expression]]s.@@@@1@13@@danf@17-8-2009
10631210@unknown@formal@none@1@S@* The [[Cyc]] project has its own ontology language called [[CycL]], based on [[first-order predicate calculus]] with some higher-order extensions.@@@@1@20@@danf@17-8-2009
10631220@unknown@formal@none@1@S@* [[Rule Interchange Format]] (RIF) and [[F-Logic]] combine ontologies and rules.@@@@1@11@@danf@17-8-2009
10631230@unknown@formal@none@1@S@* The [[Gellish]] language includes rules for its own extension and thus integrates an ontology with an ontology language.@@@@1@19@@danf@17-8-2009
10631240@unknown@formal@none@1@S@== Relation to the philosophical term ==@@@@1@7@@danf@17-8-2009
10631250@unknown@formal@none@1@S@The term ''ontology'' has its origin in [[ontology|philosophy]], where it is the name of one fundamental branch of [[metaphysics]], concerned with analyzing various types or modes of ''existence'', often with special attention to the relations between particulars and universals, between intrinsic and extrinsic properties, and between essence and existence.@@@@1@49@@danf@17-8-2009
10631260@unknown@formal@none@1@S@According to [[Tom Gruber]] at [[Stanford University]], the meaning of ''ontology'' in the context of computer science is “a description of the concepts and relationships that can exist for an [[Software agent|agent]] or a community of agents.”@@@@1@37@@danf@17-8-2009
10631270@unknown@formal@none@1@S@He goes on to specify that an ontology is generally written, “as a set of definitions of formal vocabulary.”@@@@1@19@@danf@17-8-2009
10631280@unknown@formal@none@1@S@What ontology has in common in both computer science and philosophy is the representation of entities, ideas, and events, along with their properties and relations, according to a system of categories.@@@@1@31@@danf@17-8-2009
10631290@unknown@formal@none@1@S@In both fields, one finds considerable work on problems of ontological relativity (e.g. [[Quine]] and [[Kripke]] in philosophy, [[John F. Sowa|Sowa]] and [[Nicola Guarino|Guarino]] in computer science (Top-level ontological categories.@@@@1@30@@danf@17-8-2009
10631310@unknown@formal@none@1@S@By: Sowa, John F.@@@@1@4@@danf@17-8-2009
10631320@unknown@formal@none@1@S@In International Journal of Human-Computer Studies, v. 43 (November/December 1995) p. 669-85.), and debates concerning whether a normative ontology is viable (e.g. debates over [[foundationalism]] in philosophy, debates over the [[Cyc]] project in AI).@@@@1@34@@danf@17-8-2009
10631330@unknown@formal@none@1@S@Differences between the two are largely matters of focus.@@@@1@9@@danf@17-8-2009
10631340@unknown@formal@none@1@S@Philosophers are less concerned with establishing fixed, controlled vocabularies than are researchers in computer science, while computer scientists are less involved in discussions of first principles (such as debating whether there are such things as fixed essences, or whether entities must be ontologically more primary than processes).@@@@1@47@@danf@17-8-2009
10631350@unknown@formal@none@1@S@During the second half of the 20th century, philosophers extensively debated the possible methods or approaches to building ontologies, without actually ''building'' any very elaborate ontologies themselves.@@@@1@27@@danf@17-8-2009
10631360@unknown@formal@none@1@S@By contrast, computer scientists were building some large and robust ontologies (such as [[WordNet]] and [[Cyc]]) with comparatively little debate over ''how'' they were built.@@@@1@25@@danf@17-8-2009
10631370@unknown@formal@none@1@S@In the early years of the 21st century, the interdisciplinary project of [[cognitive science]] has been bringing the two circles of scholars closer together.@@@@1@24@@danf@17-8-2009
10631380@unknown@formal@none@1@S@For example, there is talk of a "computational turn in philosophy" which includes philosophers analyzing the formal ontologies of computer science (sometimes even working directly with the software), while researchers in computer science have been making more references to those philosophers who work on ontology (sometimes with direct consequences for their methods).@@@@1@52@@danf@17-8-2009
10631390@unknown@formal@none@1@S@Still, many scholars in both fields are uninvolved in this trend of cognitive science, and continue to work independently of one another, pursuing separately their different concerns.@@@@1@27@@danf@17-8-2009
10631400@unknown@formal@none@1@S@==Resources==@@@@1@1@@danf@17-8-2009
10631410@unknown@formal@none@1@S@===Examples of published ontologies ===@@@@1@5@@danf@17-8-2009
10631420@unknown@formal@none@1@S@* [[Dublin Core]], a simple ontology for documents and publishing.@@@@1@10@@danf@17-8-2009
10631430@unknown@formal@none@1@S@* [[Cyc]] for formal representation of the universe of discourse.@@@@1@10@@danf@17-8-2009
10631440@unknown@formal@none@1@S@* [[Suggested Upper Merged Ontology]], which is a formal upper ontology@@@@1@11@@danf@17-8-2009
10631450@unknown@formal@none@1@S@* [http://www.ifomis.org/bfo/ Basic Formal Ontology (BFO)], a formal upper ontology designed to support scientific research@@@@1@15@@danf@17-8-2009
10631460@unknown@formal@none@1@S@* [[Gellish English dictionary]], an ontology that includes a dictionary and taxonomy that includes an upper ontology and a lower ontology that focusses on industrial and business applications in engineering, technology and procurement.@@@@1@33@@danf@17-8-2009
10631470@unknown@formal@none@1@S@* [http://www.fb10.uni-bremen.de/anglistik/langpro/webspace/jb/gum/index.htm Generalized Upper Model], a linguistically-motivated ontology for mediating between clients systems and natural language technology@@@@1@17@@danf@17-8-2009
10631480@unknown@formal@none@1@S@* [[WordNet]] Lexical reference system@@@@1@5@@danf@17-8-2009
10631490@unknown@formal@none@1@S@* [[OBO Foundry]]: a suite of interoperable reference ontologies in biomedicine.@@@@1@11@@danf@17-8-2009
10631500@unknown@formal@none@1@S@* The [[Ontology for Biomedical Investigations]] is an open access, integrated ontology for the description of biological and clinical investigations.@@@@1@20@@danf@17-8-2009
10631510@unknown@formal@none@1@S@* [http://colab.cim3.net/file/work/SICoP/ontac/COSMO/ COSMO]: An OWL ontology that is a merger of the basic elements of the OpenCyc and SUMO ontologies, with additional elements.@@@@1@23@@danf@17-8-2009
10631520@unknown@formal@none@1@S@* [[Gene Ontology]] for [[genomics]]@@@@1@5@@danf@17-8-2009
10631530@unknown@formal@none@1@S@* [http://pir.georgetown.edu/pro/ PRO], the Protein Ontology of the Protein Information Resource, Georgetown University.@@@@1@13@@danf@17-8-2009
10631540@unknown@formal@none@1@S@* [http://proteinontology.info/ Protein Ontology] for [[proteomics]]@@@@1@6@@danf@17-8-2009
10631550@unknown@formal@none@1@S@* [http://sig.biostr.washington.edu/projects/fm/AboutFM.html Foundational Model of Anatomy] for human anatomy@@@@1@9@@danf@17-8-2009
10631560@unknown@formal@none@1@S@* [[SBO]], the Systems Biology Ontology, for computational models in biology@@@@1@11@@danf@17-8-2009
10631570@unknown@formal@none@1@S@* [http://www.plantontology.org/ Plant Ontology] for plant structures and growth/development stages, etc.@@@@1@11@@danf@17-8-2009
10631580@unknown@formal@none@1@S@* [[CIDOC|CIDOC CRM]] (Conceptual Reference Model) - an ontology for "[[cultural heritage]] information".@@@@1@13@@danf@17-8-2009
10631590@unknown@formal@none@1@S@* [http://www.linguistics-ontology.org/gold.html GOLD ] ('''G'''eneral '''O'''ntology for [[descriptive linguistics|'''L'''inguistic '''D'''escription ]])@@@@1@11@@danf@17-8-2009
10631600@unknown@formal@none@1@S@* [http://www.landcglobal.com/pages/linkbase.php Linkbase] A formal representation of the biomedical domain, founded upon [http://www.ifomis.org/bfo/ Basic Formal Ontology (BFO)].@@@@1@17@@danf@17-8-2009
10631610@unknown@formal@none@1@S@* [http://www.loa-cnr.it/Ontologies.html Foundational, Core and Linguistic Ontologies]@@@@1@7@@danf@17-8-2009
10631620@unknown@formal@none@1@S@* [[ThoughtTreasure]] ontology@@@@1@3@@danf@17-8-2009
10631630@unknown@formal@none@1@S@* [[LPL]] Lawson Pattern Language@@@@1@5@@danf@17-8-2009
10631640@unknown@formal@none@1@S@* [[TIME-ITEM]] Topics for Indexing Medical Education@@@@1@7@@danf@17-8-2009
10631650@unknown@formal@none@1@S@* [[POPE]] Purdue Ontology for Pharmaceutical Engineering@@@@1@7@@danf@17-8-2009
10631660@unknown@formal@none@1@S@* [[IDEAS Group]] A formal ontology for enterprise architecture being developed by the Australian, Canadian, UK and U.S. Defence Depts. [http://www.ideasgroup.org The IDEAS Group Website]@@@@1@25@@danf@17-8-2009
10631670@unknown@formal@none@1@S@* [http://www.eden-study.org/articles/2007/problems-ontology-programs_ao.pdf program abstraction taxonomy]@@@@1@5@@danf@17-8-2009
10631680@unknown@formal@none@1@S@* [http://sweet.jpl.nasa.gov/ SWEET] Semantic Web for Earth and Environmental Terminology@@@@1@10@@danf@17-8-2009
10631690@unknown@formal@none@1@S@* [http://www.cellcycleontology.org/ CCO] The Cell-Cycle Ontology is an application ontology that represents the cell cycle@@@@1@15@@danf@17-8-2009
10631700@unknown@formal@none@1@S@===Ontology libraries===@@@@1@2@@danf@17-8-2009
10631710@unknown@formal@none@1@S@The development of ontologies for the Web has led to the apparition of services providing lists or directories of ontologies with search facility.@@@@1@23@@danf@17-8-2009
10631720@unknown@formal@none@1@S@Such directories have been called ontology libraries.@@@@1@7@@danf@17-8-2009
10631730@unknown@formal@none@1@S@The following are static libraries of human-selected ontologies.@@@@1@8@@danf@17-8-2009
10631740@unknown@formal@none@1@S@* The [http://www.daml.org/ontologies/ DAML Ontology Library] maintains a legacy of ontologies in DAML.@@@@1@13@@danf@17-8-2009
10631750@unknown@formal@none@1@S@* The [http://protegewiki.stanford.edu/index.php/Protege_Ontology_Library Protege Ontology Library] contains a set of owl, Frame-based and other format ontologies.@@@@1@16@@danf@17-8-2009
10631760@unknown@formal@none@1@S@* [http://www.schemaweb.info/ SchemaWeb] is a directory of RDF schemata expressed in RDFS, OWL and DAML+OIL.@@@@1@15@@danf@17-8-2009
10631770@unknown@formal@none@1@S@The following are both directories and search engines.@@@@1@8@@danf@17-8-2009
10631780@unknown@formal@none@1@S@They include crawlers searching the Web for well-formed ontologies.@@@@1@9@@danf@17-8-2009
10631790@unknown@formal@none@1@S@* [[Swoogle]] is a directory and search engine for all RDF resources available on the Web, including ontologies.@@@@1@18@@danf@17-8-2009
10631800@unknown@formal@none@1@S@* The [http://olp.dfki.de/OntoSelect/ OntoSelect] Ontology Library offers similar services for RDF/S, DAML and OWL ontologies.@@@@1@15@@danf@17-8-2009
10631810@unknown@formal@none@1@S@* [http://www.w3.org/2004/ontaria/ Ontaria] is a "searchable and browsable directory of semantic web data", with a focus on RDF vocabularies with OWL ontologies.@@@@1@22@@danf@17-8-2009
10631820@unknown@formal@none@1@S@* The [http://www.obofoundry.org/ OBO Foundry / Bioportal]is a suite of interoperable reference ontologies in biology and biomedicine.@@@@1@17@@danf@17-8-2009