Abstract: The task of word sense disambiguation is to make computers choose the correct sense of an ambiguous word in a given context. It is important for problems in natural language processing, such as information retrieval, machine translation, text classification and automatic summarization. In this paper, a new method of word sense disambiguation is proposed, where syntactic information is introduced. The parsing tree of its context including the ambiguous word is built. Disambiguation features are extracted including parsing information, part of speech and word information. The Bayesian model is used to build word sense disambiguation classifier. Experimental results show that accuracy rate of disambiguation is improved and arrives at 65%.

Key words: word sense disambiguation, syntactic information, disambiguation features, Bayesian model

摘要： 词义消歧要解决如何让计算机理解多义词在上下文中的具体含义，对信息检索、机器翻译、文本分类和自动文摘等自然语言处理问题有着十分重要的作用。通过引入句法信息，提出了一种新的词义消歧方法。构造歧义词汇上下文的句法树，提取句法信息、词性信息和词形信息作为消歧特征。利用贝叶斯模型来建立词义消歧分类器，并将其应用到测试数据集上。实验结果表明：消歧的准确率有所提升，达到了65%。

关键词: 词义消歧, 句法信息, 消歧特征, 贝叶斯模型