基于联合法选取特征的产品评论情感分类研究
|
摘要
随着互联网中信息资源不断膨胀,如何从海量的产品评论文本信息中获得对某一产品真实的、客观的评价,已成为一个亟待解决的问题。针对传统特征量选择方法的不足,该文采用了联合特征选取的方法,并构建了基于支持向量机的分类器,有效地实现了对产品评论文本情感倾向性的分类。对网络中获取的大量评论信息进行分析和研究,结果表明,提出的方法有效地克服了低频词中噪音词的干扰,提高了文本表示质量,改善了分类的效果。
With the ceaseless inflation of information resources on the Internet,how to get a real and objective evaluation about a product from huge amounts of product reviews information has becoming an urgent problem which should be solved. Aiming at the shortages of traditional feature selection methods,this paper has adopted combined method to select features,and constructed classifier based on SVM,which realized the emotion tendentiousness classification of product reviews texts effectively. Through analysis and research on massive comment information obtained from the Internet,it turns out that the method proposed in this paper conquers noise words' disturbance existing in low frequency words with effect,enhanced the quality of text representation,and improved the result of classification.
With the ceaseless inflation of information resources on the Internet,how to get a real and objective evaluation about a product from huge amounts of product reviews information has becoming an urgent problem which should be solved. Aiming at the shortages of traditional feature selection methods,this paper has adopted combined method to select features,and constructed classifier based on SVM,which realized the emotion tendentiousness classification of product reviews texts effectively. Through analysis and research on massive comment information obtained from the Internet,it turns out that the method proposed in this paper conquers noise words' disturbance existing in low frequency words with effect,enhanced the quality of text representation,and improved the result of classification.
引文
[1]余珍芝.中文网络产品评论的情感分析关键技术研究[D].杭州:杭州电子科技大学,2011.
[2]Pang Bo,Lee Lillian,Vaithyanathan et al.Thumbs up Sentiment classification using machine learning techniques[C].Proceedings of the Conference on Empirical Methods in Natural Language Processing,2002:79-86.
[3]Turney P D.Thumbs up or thumbs down:semantic orientation applied to unsupervised classification of reviews[C].Meeting on Association for Computational Linguistics.Association for Computational Linguistics,2002:417-424.
[4]唐焕玲.文本分类方法[M].北京:电子工业出版社,2013:1-41.
[5]党蕾.中文文本多粒度情感分类计算的研究[D].西安:西北大学,2010.
[6]张小倩.情感极性转移现象研究及应用[D].苏州:苏州大学,2012.
[7]王作飞.基于粗糙集的启发式属性约简特征选择方法研究[D].郑州:郑州大学,2011.
[8]徐琳宏,林鸿飞,赵晶.情感语料库的构建和分析[J].中文信息学报,2008,22(1):116-122.
[9]张迪.基于跨领域分类学习的产品评论情感分析[D].上海:上海交通大学,2011.
[10]刘鸿宇,赵妍妍,秦兵,等.评价对象抽取及其倾向性分析[J].中文信息学报,2010:84-88.
[12]杨彦闯,杨炳儒,张克君.基于联合提取特征的粗糙集文本分类技术研究[J].计算机应用研究,2007,24(7):97-98.
[13]黄高辉,姚天昉,刘全升.基于CRF算法的汉语比较句识别和关系抽取[J].计算机应用研究,2010,27(6):2061-2064.
[14]宋鸿彦,刘军,姚天昉,等.汉语意见型主观性文本标注语料库的构建[J].中文信息学报,2009,23(2):123-128.
[15]林培金.基于领域本体的语义合成研究及应用[D].南京:南京邮电大学,2013.
[2]Pang Bo,Lee Lillian,Vaithyanathan et al.Thumbs up Sentiment classification using machine learning techniques[C].Proceedings of the Conference on Empirical Methods in Natural Language Processing,2002:79-86.
[3]Turney P D.Thumbs up or thumbs down:semantic orientation applied to unsupervised classification of reviews[C].Meeting on Association for Computational Linguistics.Association for Computational Linguistics,2002:417-424.
[4]唐焕玲.文本分类方法[M].北京:电子工业出版社,2013:1-41.
[5]党蕾.中文文本多粒度情感分类计算的研究[D].西安:西北大学,2010.
[6]张小倩.情感极性转移现象研究及应用[D].苏州:苏州大学,2012.
[7]王作飞.基于粗糙集的启发式属性约简特征选择方法研究[D].郑州:郑州大学,2011.
[8]徐琳宏,林鸿飞,赵晶.情感语料库的构建和分析[J].中文信息学报,2008,22(1):116-122.
[9]张迪.基于跨领域分类学习的产品评论情感分析[D].上海:上海交通大学,2011.
[10]刘鸿宇,赵妍妍,秦兵,等.评价对象抽取及其倾向性分析[J].中文信息学报,2010:84-88.
[12]杨彦闯,杨炳儒,张克君.基于联合提取特征的粗糙集文本分类技术研究[J].计算机应用研究,2007,24(7):97-98.
[13]黄高辉,姚天昉,刘全升.基于CRF算法的汉语比较句识别和关系抽取[J].计算机应用研究,2010,27(6):2061-2064.
[14]宋鸿彦,刘军,姚天昉,等.汉语意见型主观性文本标注语料库的构建[J].中文信息学报,2009,23(2):123-128.
[15]林培金.基于领域本体的语义合成研究及应用[D].南京:南京邮电大学,2013.