基于统计的汉语词义消歧研究

Study on Chinese Word Sense Disambiguation Based on Statistics

词义消歧一直是信息检索领域的关键问题和难点之一。统计学方法以其良好的词义消歧效果逐渐占据主流地位。文章结合义类词典HowNet,分别采用隐马尔可夫模型,贝叶斯模型,基于依存关系改进贝叶斯模型对大规模真实文本进行了词义消歧研究,其中基于依存关系改进贝叶斯模型的词义消歧效果最好。

Word sense disambiguation has always been a key problem and one of the difficult points in information retrieval. Statistics have good effect on word sense disambiguation,and gradually takes up the mainstream status. In this paper,large-scale real texts are researched with HowNet, respectively applying Hidden Markov Model, Naive Bayes Modei,Bayes Model and Dependency Parsing, among which Bayes Model and Dependency Parsing have the best effect.