Semantic Clustering: an Attempt to Identify Multiword Expressions in Bengali

One of the key issues in both natural language understanding and generation is the appropriate processing of Multiword Expressions (MWEs). MWE can be defined as a semantic issue of a phrase where the meaning of the phrase may not be obtained from its constituents in a straightforward manner. This paper presents an approach of identifying bigram noun-noun MWEs from a medium-size Bengali corpus by clustering the semantically related nouns and incorporating a vector space model for similarity measurement. Additional inclusion of the English WordNet::Similarity module also improves the results considerably. The present approach also contributes to locate clusters of the synonymous noun words present in a document. Experimental results draw a satisfactory conclusion after analyzing the Precision, Recall and F-score values.