The input text is first passed through a format fil­ter, whose task is to parse the input XML file. This is followed by a preprocessing step. Each instance passed to the preprocessing stage is first segmented into words, and then all compound words are iden­tified. Any sequence of words known to be a com­pound in WordNet is combined into a single entity.

Although each input instance consists of a large number of words, only a few of these are likely to be useful for disambiguating the target word. We use the context selection algorithm to select a subset of the context words to be used for sense selection. By removing the unimportant words, the computa­tional complexity of the algorithm is reduced.

In this work, we use the NearestWords context selection algorithm. This algorithm algorithm se­lects 2n + 1 content words surrounding the target word (including the target word) as the context. A stop list is used to identify closed-class non-content words. Additionally, any word not found in Word­Net is also discarded. The algorithm then selects n content words before and n content words follow­ing the target word, and passes this unordered set of 2n + 1 words to the Sense Selection module.

The sense selection module takes the set of words output by the context selection module, one ofwhich is the target word to be disambiguated. For each of the words in this set, it retrieves a list of senses from WordNet, based on which it determines the intended sense of the target word.

The package provides two main algorithms for Sense Selection: the local and the global algorithms, as described in previous work (Banerjee and Peder-sen, 2002; Patwardhan et al., 2003). In this work, we use the local algorithm, which is faster and was shown to perform as well as the global algorithm.

The local sense selection algorithm measures the semantic relatedness of each sense of the target word with the senses of the words in the context, and se­lects that sense of the target word which is most re­lated to the context word-senses. Given the 2n + 1 context words, the system scores each sense of the target word. Suppose the target word t has T senses, enumerated as t\, t2,..., tT. Also, suppose w\, w2, ..., w2n are the words in the context of t, each hav­ing W\, W2,..., W2n senses, respectively. Then for each ti a score is computed as where wjk is the kth sense of word Wj. The sense tiof target word t with the highest score is selected as the intended sense of the target word.

The relatedness between two word senses is com­puted using a measure of semantic relatedness de­fined in the WordNet::Similarity software package (Pedersen et al., 2004), which is a suite of Perl mod­ules implementing a number WordNet-based mea­sures of semantic relatedness. For this work, we used the Context Vector measure (Patwardhan and Pedersen, 2006). The relatedness of concepts is computed based on word co-occurrence statistics derived from WordNet glosses. Given two WordNet senses, this module returns a score between 0 and 1, indicating the relatedness of the two senses.

Our system relies on WordNet as its sense inven­tory. However, this task used OntoNotes (Hovy et al., 2006) as the sense inventory. OntoNotes word senses are groupings of similar WordNet senses. Thus, we used the training data answer key to gen­erate a mapping between the OntoNotes senses of the given lexical elements and their corresponding WordNet senses. We had to manually create the mappings for some of the WordNet senses, which had no corresponding OntoNotes senses. The sense selection algorithm performed all of its computa­tions with respect to the WordNet senses, and finally the OntoNotes sense corresponding to the selected WordNet sense of the target word was output as the

Instance

answer for each instance.