否定信息识别研究综述  

Review on negation recognition

作　　者：唐凡[1] 伍星[1] TANG Fan;WU Xing(College of Computer Science, Chongqing University, Chongqing 400044, China)

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《计算机工程与应用》2017年第12期1-8,共8页Computer Engineering and Applications

基　　金：重庆市研究生科研创新项目(No.CYS15031)

摘　　要：自然语言中否定是一个普遍且复杂的现象,其能反转命题的真值、情感及态度的极性、观点的立场等。否定信息识别不仅在信息检索、情感分析、文本挖掘、信息抽取等领域中具有重要意义,也对文本的深层语义理解具有促进作用。对否定信息识别研究进行了深入分析,介绍了否定信息识别的任务框架,并对否定触发词检测、否定覆盖域界定、否定焦点识别三个子任务的解决思路及方法进行了详细阐述,最后总结并指出了否定信息识别研究中存在的问题及新方向。In natural language,negation is a common and complex phenomenon,it can inverse the truth value of proposition,the polarity of sentiments and attitude,the position of opinion and so on.Negation recognition plays an importantrole in many fields such as information retrieval,sentiment analysis,text mining,information extraction.It also has a positiveeffect on a further step towards text understanding.A survey of negation recognition is analyzed.This paper mainlyintroduces the framework of negation recognition task,and elaborates the solution of the task that includes three subtasks:negation cues detection,negation scope resolution and negation focus detection.Finally the problems of the current researchand the future work directions are pointed out.

关 键 词：否定信息 触发词检测 覆盖域界定 焦点识别 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]