Attribute knowledge mining for Chinese word sense disambiguation

Word sense disambiguation is a technology of judging the specific semantic of polysemous words in the specific context. It is meaningful for the applications of natural language processing. This paper introduces the attribute knowledge into word sense disambiguation task. Every sense of the polysemous words can be described by the different attribute sets. These attributes can be viewed as a kind of context features. The attribute knowledge bases are built for every polysemous word, and employed into the Naive Bayes classifier and Maximum Entropy classifier as a dimension feature to judge the specific semantic of polysemous words in the specific context. The experimental results show that this method can effectively improve the accuracy of Chinese word sense disambiguation.