基于语义和句法依存特征的评论对象抽取研究  

Opinion Target Extraction Based on Semantic and Syntactic Dependency

作　　者：张志远[1] 赵越[1] ZHANG Zhiyuan;ZHAO Yue(School of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,Chin)

机构地区：[1]中国民航大学计算机学院,天津300300

出　　处：《中文信息学报》2018年第6期80-87,97,共9页Journal of Chinese Information Processing

基　　金：国家自然基金民航联合基金(U1633110);中央高校基本科研业务费专项基金(3122016D021)

摘　　要：评论对象抽取是情感分析的重要研究内容。基于语义词典,从评论对象的类别视角出发,运用语义相似度和相关度计算方法,该文提出用于评价对象抽取的七种新的语义特征。评价对象和评价词之间通常存在句法依存关系,并且评价词往往带有情感倾向,将句法依存分析和评价词识别结合,提出句法情感依存特征抽取方法,忽略无情感词和微情感词的句法依存关系,提高评价对象抽取的准确率。使用条件随机场模型,在SEMEVAL比赛的三个领域数据集上进行实验,新的语义特征和句法情感依存特征组合的F1分数比SEMEVAL比赛限制性系统最好成绩平均高3.78%,比非限制性系统最好成绩平均高2%,证明了所提特征的有效性。Opinion target extraction is an important task of sentiment analysis. Based on a semantic dictionary, this paper proposes seven semantic features of opinion targets in relation to their categories via the semantic similarity and relevance computation. Since there are exist syntactic dependency between the opinion targets and opinion words, this paper further presents the extraction method of sentiment syntactic dependency features, ignoring those objective words or micro sentiment words to improve the accuracy. In the experiments on three datasets of SEME- VAL,the combination of new semantic features and sentiment syntactic dependency features enable the CRFs a F1 score of 3.78 points higher than the SEMEVAL＇s best score for constrained systems,and 2 points higher for uncon strained systems.

关 键 词：评价对象抽取 条件随机场 语义特征 句法依存关系 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]