A Comparison of Bag of Words and Syntax-based Approaches for Word Categorization

This paper will examine the aptness of various word space models for the task of word categorization, as defined by the lexical semantics workshop at ESSLLI 2008. Three word clustering tasks will be examined: concrete noun categorization, concrete/abstract noun discrimination, and verb categorization. The main focus will be on the difference between bag of words models and syntax-based models. Both approaches will be evaluated with regard to the three tasks, and differences between the clustering solutions will be pointed out.