基于贝叶斯模型的复句关系词自动识别与规则挖掘.pdfVIP

Vo l. 42 No.7 计算机科学 第42 卷第7 期 2015 年7 月 Computer Science July 2015 基于贝叶斯模型的复旬关系词自动识别与规则挖掘 杨进才郭凯凯沈显君胡金柱 (华中师范大学计算机学院 武汉 430079) 摘 要 复句是汉语语法的重要实体单位，关系词的自动识别是复句标识的基础，对立句的标识以及篇章的研究有重 要意义。在对汉语复句语料库进行广泛分析的基础上，从复句关系词所在的环境和关系词的组合搭配方面进行特征 的提取，对提取的特征进行形式化描述。采用互信息和信息增益相结合的方式进行特征选择以及冗余特拉的消除川史 用贝叶斯模型对特征集令进行训练和测试;4守基于统计过程的结果转化为规则.形成规则库，并根据规则进行关系词 自动识别。实验结果显示，本方法获得了较高的识JI}j正确牟，具有可行性和有效险。 关键词 复句关系祠，贝叶其斤，规则，自动标识 中图法分类号 TP391 文献标识码 A ∞1 10. 11896/ j. issn. 1002-137X. 2015. 7. 062 Automatic Identification and Rule Mining for Relati佣Words of Chin臼e Compound Sentences Based on Bay臼ian M创el YANG Jin-cai GUO Kai-kai SHEN Xian-jun HU Jin-zhu Sch∞lof 臼mputer Science .Huazhong Nom咀1 University. Wuhan 430079. China) Abstract The compound sentence is an important unit of the Chinese sentence and its annotation is important to the re- search on comprehending Chinese texts. Identification of relation words is the basis of compound sentence annotation. Based on a comprehensive analysis of Chinese compounds corpus ,this paper extracted features of relation words from their context and collocation. Those features are described in formulas. A combination of mutual information with infor- rnation gains is used for selecting features and eliminating redundant features. The Bayesian model is used for training and testing feature sets. Rules are created from the statistics results ,and rule base is configured with rules. which are used for autornatic identification of relation words. The experimental results show that our method obtains a high acc