作为文本内容理解的媒介与载体,词汇语义知识库已被广泛应用于信息检索、信息提取、问答系统、自动文摘等方面,成为自然语言处理不可或缺的基础资源。本文介绍词汇语义知识库研究与开发的现状,重点分析了WordNet、SinicaBOW、HowNet及CCD等具有代表性的词汇语义知识库的具体情况。在此基础上,盘点各种需求和解决方案,提出词汇语义知识库研究面临新的挑战和机遇,即本体化和多语化的大趋势,它们将从不同方面弥补词汇语义知识库在知识共享和知识交流上的不足,使其更好地为自然语言处理服务。本文最后探讨了词汇语义知识库未来发展中可能存在的问题和新的课题。
As the semantic knowledge carrier for text understanding by computers, lexical semantic knowledge bases have been widely used in natural language processing such as machine translation, infonnation retrieval, etc. We give a detailed analysis of the state-of-the-art of and tendencies in them, which are referred to as ontologieahzation and multilinguahzation respectively. These two developments could remedy of the defects of knowledge share and knowledge communication of lexical semantic knowledge bases. The existing problems and future directions are at last put forward.