Efficient implementation of dot product and bounded Jiang-Conrath similarity
svectors is a free implementation of two forms of computing word similarity. Specifically, it computes the dot product on sparse vectors (for the cosine similarity measure) and a bounded variant of the Jiang-Conrath similarity measure on the WordNet taxonomy. If you don’t know what these are, you probably don’t need this software. In particular, svectors will not simply compute word similarity for you. It’s mainly a tool for researchers in computational linguistics, and you need to feed it with appropriate data.