The Computation of Chinese Word Similarity Based on Large Scale Corpus
Zeng Sai · Journal of Zhongyuan University of Technology · 2010
Word similarity is a fundamental problem in natural language processing.Chinese word similarity was computed based on a large scale corpus.First,a platform for computing the word similarity was implemented.This platform is easy for researchers to combine all kinds of algorithms to obtain the word similarity.Meanwhile,the additional corpus can be proceeded in an incremental computation way.At last,the Euclidean measure and probability measure were evaluated by gold standard which was created by human.