Word Sense Disambiguation Based on Vicarious Words

This paper presents the concept of vicarious words and develops a new unsupervised Chinese word sense disambiguation method. This method, after statistical learning from the vicarious words, realizes unsupervised word sense disambiguation by calculating mutual information to measure the degree of collocation information between the ambiguous words and their context. In our experiment, we test ten real ambiguous words and get the highest accuracy of 97%. The mean accuracy was 88.52%. The experimental results have proved the feasibility of this method.