Study on Word Sense Disambiguation Knowledge Base Based on Multi-Sources

Although there are many resources used in Natural Language Process, a specialized knowledge base used in word sense disambiguation(WSD) is still a shortage. By extracting knowledge from different resources and using them, we can improve the accuracy rate of word sense disambiguation. In this article, we use different methods existed to extract properties from The Grammatical Knowledge-base of Contemporary Chinese(GKB), HowNet, The Word-Sense Tagging Corpus (STC) and The Semantic Knowledge-base of Contemporary Chinese(SKCC) to build a specific knowledge base, which can help us with Chinese word sense disambiguation.