基于依存关系规则的汉语复句关系词自动识别  

Research on automatic recognition of relational word in Chinese compound sentences based on dependency relation rule

作　　者：杨进才[1] 涂馨丹 沈显君[1] 胡金柱[1] Yang Jincai;Tu Xindan;Shen Xianjun;Hu Jinzhu(School of Computer Science,Central China Normal University,Wuhan 430079,Chin)

机构地区：[1]华中师范大学计算机学院,武汉430079

出　　处：《计算机应用研究》2018年第6期1756-1760,共5页Application Research of Computers

基　　金：国家社会科学基金资助项目(14BYY093)

摘　　要：目前对关系词的自动识别主要使用分词的字面特征,较少涉及语法特征的分析。在基于字面特征的关系词搭配的约束条件基础上,研究基于依存关系的约束条件。利用依存语法中依存关系的五条公理分析汉语复句中关系词搭配的依存关系,从中总结出七类约束条件,并把这七类约束条件形成依存关系规则加入规则库,运用字面特征与语法特征相结合的规则自动识别汉语复句关系词。实验结果表明,对关系词的识别率可以达到95.77%。Current research about the recognition of the relational word is mainly concentrated in literal features,the research refer to syntax analysis was relatively minority. Based on the constraint conditions of Chinese complex sentence relational words＇ collocation which was on account of literal characteristics,this paper studied the constraint conditions research which were based on the dependency relation,it used five axioms of dependency grammar for analyzing relation words＇ dependency relation on Chinese complex sentence,and summarized seven kinds of constraints,then formalized these constraints into dependency relation rule to join the rule base,it used rule which combined literal features with grammar features to automatic identify relational word of Chinese complex sentence. The experimental results show that the rate of the relational words ＇recognition is 95. 77 percent.

关 键 词：关系词搭配 依存关系 约束条件 自动识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]