基于双层HHMM的产品评论特征和情感分类

被引:3
作者
张磊
李梦诗
陈黎
黎红友
李志蜀
彭舰
机构
[1] 四川大学计算机学院
关键词
Web数据挖掘; 特征情感分类; 标注规则; 双层HHMM;
D O I
10.15961/j.jsuese.2013.02.015
中图分类号
TP391.1 [文字信息处理];
学科分类号
摘要
近年来,中文产品评论的特征情感分类是Web数据挖掘的重要研究内容之一。提出了一套完整的产品命名实体、特征词、情感词以及边界的标注规则,设计了多层次的混合标签模式;提出了双层HHMM(层级隐马尔科夫模型)结构,将词形标注和词性标注的特点进行融合;提出了基于词形标注的HHMM-1算法和基于词性标注的HHMM-2算法,实现复杂短语的自动标注。实验证明,双层HHMM模型起到了互补的作用,模型的查全率和F-score值均有较大提高。
引用
收藏
页码:94 / 102
页数:9
相关论文
共 26 条
[1]   基于中文文本主题跟踪的网络信息分析 [J].
王科 ;
刘渊 ;
罗万伯 ;
高行宇 ;
高常波 .
四川大学学报(工程科学版), 2004, (01) :114-118
[2]   Product named entity recognition in Chinese text [J].
Zhao, Jun ;
Liu, Feifan .
LANGUAGE RESOURCES AND EVALUATION, 2008, 42 (02) :197-217
[3]  
The Hierarchical Hidden Markov Model: Analysis and Applications[J] . Shai Fine,Yoram Singer,Naftali Tishby. &nbspMachine Learning . 1998 (1)
[4]  
Sentiment analyzer: Extracting sentiments about a giventopic using natural language processing techniques. J. Yi,T. Nasukawa,R. C. Bunescu, et al. The3rd IEEE International Conference on DataMining . 2003
[5]  
A Named Entity Recognizer for Danish. Bick,Eckhard. Proc. 4th International Conf. on Language Resources and Evaluation(LREC2004) . 2004
[6]  
Learning Hidden Markov Model Structure for Information Extraction. Kristie Seymore,Andrew McCallum,Ronald Rosenfeld. AAAI-99 Workshop on Machine Learning for Information Extraction . 1999
[7]  
Mining and summarizing customer reviews. Minqing Hu,Bing Liu. Proceedings of the 10th ACM SIGKDD international conference on Knowledge discovery and data mining . 2004
[8]  
Opinion Mining and Sentiment Analysis. Pang B.,L Lee. Foundations and Trends in Information Retrieval . 2008
[9]  
Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data. Liu B. . 2006
[10]  
LTP:a chineselanguage technology platform. Che Wanxian,Li Zhenghua,Liu Ting. Proceedings of theColing 2010:Demonstrations . 2010