基于级联模型的中文情感要素抽取

doi:10.3969/j.issn.0372-2112.2016.10.025

电子学报 ›› 2016, Vol. 44 ›› Issue (10): 2459-2465.DOI: 10.3969/j.issn.0372-2112.2016.10.025

基于级联模型的中文情感要素抽取

王亚珅, 黄河燕, 冯冲, 刘全超

北京理工大学计算机学院北京市海量语言信息处理与云计算应用工程技术研究中心, 北京 100081

收稿日期:2015-02-11 修回日期:2015-06-26 出版日期:2016-10-25 发布日期:2016-10-25
通讯作者: 黄河燕
作者简介:王亚珅,男,1989年出生,北京理工大学计算机科学与技术专业博士研究生,主要研究领域为社交网络分析和信息检索.E-mail:yswang@bit.edu.cn
基金资助:
国家重点基础研究发展计划（973计划）资助项目（No.2013CB329605，No.2013CB329303）；国家自然科学基金（No.61132009，No.61201351）

Chinese Evaluation Element Extraction Based on Cascaded Model

WANG Ya-shen, HUANG He-yan, FENG-Chong, LIU Quan-chao

Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications, School of Computer, Beijing Institute of Technology, Beijing 100081, China

Received:2015-02-11 Revised:2015-06-26 Online:2016-10-25 Published:2016-10-25

摘要/Abstract

摘要：

随着社交媒体的发展及成熟，每天在互联网环境中都会产生大量的用户评论信息.抽取评价短语、评价对象和观点持有者等情感要素，已经成为了中文观点挖掘和情感分析的重要先决任务.针对中文情感要素抽取任务，本文提出了一个统计和规则相结合的级联模型，主要贡献包括：（1）针对汽车领域评论信息，构建情感要素标注语料库和相关词典；（2）对于以往研究较少关注的中文评价短语，本文详细分析阐述其定义和分类；（3）结合统计和规则，分别针对评价短语和情感要素提出级联抽取策略.实验结果充分证明了该级联模型的有效性，相比较于其它基于规则的情感要素抽取算法有效提升了召回率，同时为后续社交媒体情感分析任务提供了有力的支持.

关键词: 信息抽取, 情感要素, 评价短语, 评价对象, 观点持有者

Abstract:

With the development of social media,massive reviews are generated by users every day.The extraction of evaluation elements,including evaluation phrase,comment target and opinion holder,is an important pre-task of Chinese opinion mining and sentiment analysis.This paper proposes an efficient method for extracting Chinese evaluation elements based on cascaded model and mainly makes three contributions:(i) to implement and evaluate the method,we construct an original annotated corpus for Chinese evaluation elements of automobile;(ii) we provide specific definition and classification of Chines evaluation phrase;(iii) combing statistic method and rule-based method,we present cascaded strategy for extraction of evaluation phrase and evaluation elements,respectively.According to the experiment results,the proposed method performs well,and effectively improve the recall compared with other rule-based algorithm.Meanwhile it contributes greatly to our subsequent tasks,such as sentiment analysis of social media.

Key words: information extraction, evaluation element, evaluation phrase, comment target, opinion holder

中图分类号:

TP391.1

王亚珅, 黄河燕, 冯冲, 刘全超. 基于级联模型的中文情感要素抽取[J]. 电子学报, 2016, 44(10): 2459-2465.

WANG Ya-shen, HUANG He-yan, FENG-Chong, LIU Quan-chao. Chinese Evaluation Element Extraction Based on Cascaded Model[J]. Acta Electronica Sinica, 2016, 44(10): 2459-2465.

参考文献

[1] Pang B,Lee L.Opinion mining and sentiment analysis Foundations and trends in information retrieval[J].Foundations & Trends in Information Retrieval,2008,2(1-2):459-526.
[2] Liu B.Sentiment analysis and opinion mining[J].Synthesis Lectures on Human Language Technologies,2012,5(1):1-167.
[3] Bloom K,Garg N,Argamon S.Extracting appraisal expressions[A].Proceedings of Human Language Technologies:2007 Annual Conference of the North American Chapter of the Association for Computational Linguistics[C].New York:ACL Press,2007.308-315.
[4] 赵妍妍,秦兵,车万翔,刘挺.基于句法路径的情感评价单元识别[J].软件学报,2011,22(5):887-898. ZHAO Yan-yan,QIN Bing,CHE Wan-xiang,LIU Ting.Appraisal expression recognition based on syntactic path[J].Journal of Software,2011,22(5):887-898.(in Chinese)
[5] Whitelaw C,Garg N,Argamon S.Using appraisal groups for sentiment analysis[A].Proceedings of 14^th ACM International Conference on Information and Knowledge Management[C].New York,USA:ACM Press,2005.625-631.
[6] Nakagawa T,Kawada T,Inui K,Kurohashi S.Extracting subjective and objective evaluative expressions from the Web[A].Proceedings of 2nd International Symposium on Universal Communication[C].Osaka,Japan:IEEE Press,2008.251-258.
[7] Popescu A M,Etzioni O.Extracting product features and opinions from reviews[A].Proceedings of Human Language Technology:2005 Conference on Empirical Methods in Natural Language Processing[C].Vancouver,Canada:ACL Press,2005.339-346.
[8] 赵妍妍,秦兵,刘挺.文本情感分析[J].软件学报,2010,21(8):1834-1848. ZHAO Yan-yan,QIN Bing,LIU Ting.Sentiment analysis[J].Journal of Software,2010,21(8):1834-1848.(in Chinese)
[9] 宋锐,洪莉,林鸿飞.基于ChunkCRF的观点持有者识别及其在观点摘要中的应用[J].小型微型计算机系统,2009,30(7):1462-1466. SONG Rui,HONG Li,LIN Hong-fei.Chunk-CRF-based opinion holder identification and application to opinion summarization[J].Journal of Chinese Computer Systems,2009,30(7):1462-1466.(in Chinese)
[10] Nakagawa T,Inui K,Kurohashi S.Dependency tree-based sentiment classification using CRFs with hidden variables[A].Proceedings of Human Language Technologies:2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics[C].Los Angeles,USA:ACL Press,2010.786-794.
[11] Wang Y,Kazama J,Kawada T,Torisawa K.Chinese evaluative information analysis[A].Proceedings of 24^th International Conference on Computational Linguistics[C].Mumbai,India:ACM Press,2012.2773-2788.
[12] 侯敏,滕永林,陈毓麒.评价短语的倾向性分析研究[J].中文信息学报,2013,27(6):103-109. HOU Min,TENG Yong-Lin,CHEN Yu-qi.Research on orientation analysis of opinion phrases[J].Journal of Chinese Information Processing,2013,27(6):103-109.(in Chinese)
[13] 王荣洋,鞠久朋,李寿山,周国栋.基于CRFs的评价对象抽取特征研究[J].中文信息学报,2012,26(2):56-61. WANG Rong-yang,JU Jiu-ming,LI Shou-shan,ZHOU Guo-dong.Feature engineering for CRFs based opinion target extraction[J].Journal of Chinese Information Processing,2012,26(2):56-61.(in Chinese)
[14] Hu M Q,Liu B.Mining and summarizing customer reviews[A].Proceedings of 2004 ACM SIGKDD International Conference on Knowledge Discovery & Data Mining[C].New York:ACM Press,2004.168-177.
[15] 姚天昉,等.一个用于汉语汽车评论的意见挖掘系统[A].中国中文信息学会.中文信息处理前沿进展——中国中文信息学会二十五周年学术会议论文集[C].北京:中国中文信息学会,2006.260-281. YAO Tian-fang,et al.An opinion mining system for chinese automobile reviews[A].Proceedings of the 25^th Annual Conference of CIPS[C].Beijing:Chinese Information Processing Society of China,2006.260-281.(in Chinese)
[16] 赵巍,等.连续字符识别的级联HMM训练算法[J].计算机学报,2007,30(12):2142-2150. ZHAO Wei,et al.Cascaded HMM training algorithm for continuous character recognition[J].Chinese Journal of Computers,2007,30(12):2142-2150.(in Chinese)
[17] 李本阳,等.基于单层标注级联模型的篇章情感倾向分析[J].中文信息学报,2012,26(4):3-8+20. LI Ben-yang,et al.Single-label cascaded model for document sentiment analysis[J].Journal of Chinese Information Processing,2012,26(4):3-8+20.(in Chinese)
[18] Lafferty J,McCallum A,Pereira F.Conditional random fields:Probabilistic models for segmenting and labeling sequence data[A].Proceedings of 18^th International Conference on Machine Learning[C].Williamstown,MA,USA:ACM Press,2001.282-289.

基于级联模型的中文情感要素抽取

Chinese Evaluation Element Extraction Based on Cascaded Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	刘全超, 黄河燕, 冯冲. 面向中文微博的评价对象与评价词语联合抽取[J]. 电子学报, 2016, 44(7): 1662-1670.
[2]	薛涵, 秦兵, 刘挺. 基于Folksonomy的本体构建综述[J]. 电子学报, 2014, 42(4): 791-797.
[3]	付博, 赵世奇, 刘挺. Web查询日志研究综述[J]. 电子学报, 2013, 41(9): 1800-1808.
[4]	李维刚;刘挺;李生. 基于网络挖掘的实体关系元组自动获取[J]. 电子学报, 2007, 35(11): 2111-2116.
[5]	周顺先;林亚平;王耀南;易叶青. 基于二阶隐马尔可夫模型的文本信息抽取[J]. 电子学报, 2007, 35(11): 2226-2231.
[6]	林亚平;刘云中;周顺先;陈治平;蔡立军. 基于最大熵的隐马尔可夫模型文本信息抽取[J]. 电子学报, 2005, 33(2): 236-240.
[7]	李向阳;陆建江;张亚非. 基于竞争分类的Web信息抽取[J]. 电子学报, 2004, 32(11): 1915-1917.