基于多标记学习的汽车评论文本多性能识别

J4 ›› 2016, Vol. 38 ›› Issue (01): 188-194.

• 论文 • 上一篇

基于多标记学习的汽车评论文本多性能识别

张晶,李德玉,王素格

（山西大学计算机与信息技术学院,山西太原 030006）

收稿日期:2015-10-08 修回日期:2015-12-06 出版日期:2016-01-25 发布日期:2016-01-25
基金资助:
国家自然科学基金(61272095,61175067)；山西省科技攻关项目(2011032102702)；山西省回国留学人员科研项目(2013014)；山西省科技基础条件平台建设项目(20150910010102)

Multiple performances identification for car
review texts based on multilabel learning

ZHANG Jing,LI Deyu,WANG Suge

（School of Computer and Information Technology,Shanxi University,Taiyuan 030006,China）

Received:2015-10-08 Revised:2015-12-06 Online:2016-01-25 Published:2016-01-25

摘要/Abstract

摘要：

针对汽车产品评论文本中出现的多方面性能,提出一种基于多标记学习的汽车评论文本多方面性能识别方法。首先,结合文本挖掘方法,利用多标记文本特征选择方法选取特征,将非结构化的文本转化为结构化的多标记数据集。在此基础上,使用四种多标记分类方法,对待识别的评论文档标注一个或多个方面标记。最后,以八种多标记评价指标评估方面识别的性能。在新浪汽车评论语料上的实验表明,方面识别的子集准确率达到了95%,验证了方法的可行性。

关键词: 多标记学习, 文本处理, 汽车评论, 多方面识别

Abstract:

Aiming at the characteristics of the multiaspect performance appeared in the automotive product reviews,this paper proposed a novel method for recognizing the multiple aspects of performance about car comment text based on multilabel learning.Firstly,appropriate words were selected as features by multilabel text feature selection method combined with the text mining technology,and then,the unstructured document corpus are transformed into structured multilabel dataset.After that,we finished marking one or more aspect tags for the unrecognized comment text with four multilabel classification methods.Finally,the recognition accuracy of multiple aspects was assessed by eight multilabel evaluation metrics.On the Sina car review corpus,experimental results indicate the subset accuracy reaches up to 95%.Hence,our method was feasible for recognizing the multiple aspects of automobile reviews.

Key words: multilabel learning, text processing, car reviews, multiaspect recognition

张晶,李德玉,王素格. 基于多标记学习的汽车评论文本多性能识别[J]. J4, 2016, 38(01): 188-194.

ZHANG Jing,LI Deyu,WANG Suge. Multiple performances identification for car
review texts based on multilabel learning [J]. J4, 2016, 38(01): 188-194.

参考文献［21］

［1］	Dave K,Lawrence S,Pennock D.Minging the peanut gallery:Opinion extraction and semantic classification of product reTable 2Results of classification on eight kinds of
	multilabel evaluation metrics for four kinds of classifiers表24种分类器在8种多标记评价指标上的分类结果BRCCMLkNNRankSVMHL0008 5±0001 90008 4±0002 00143 6±0010 90060 3±0012 8SA0950 0±0012 10950 5±0012 40451 5±0029 90715 5±0063 5F10974 9±0003 30975 0±0003 40664 3±0031 70877 2±0037 8A0969 9±0005 20970 1±0005 30609 0±0031 10839 6±0044 1AP0992 9±0001 60992 9±0001 60851 7±0015 90971 6±0015 0C0644 5±0068 30645 0±0068 01344 0±0047 50745 5±0040 5RL0006 6±0001 20006 6±0001 10130 6±0011 30022 7±0007 9OE0007 5±0003 10007 5±0003 10183 6±0028 40034 5±0028 7views［C］∥ Proc of the 12th International Conference on World Wide Web （WWW）,2003:519529.
［2］	Qiu G,Liu B,Bu J,et al.Expanding domain sentiment lexicon through double propagation［C］∥Proc of the 21st International Joint Conference on Artificial Intelligence（ IJCAI）,2009:11991204.
［3］	Huang S,Peng W,Li J,et al.Sentiment and topic analysis on social media:A multitask multilabel classification approach［C］∥Proc of the 5th Annual ACM Web Science Conference （WebSci）,2013:172181.
［4］	Shuhua Monica L,JiunHung C.A multilabel classification based approach for sentiment classification［J］.Expert Systems with Applications,2015,42（3）:10831093.
［5］	Katakis I,Tsoumakas G,Vlahavas I.Multilabel text classification for automated tag suggestion［C］∥Proc of the ECML/PKDD 2008 Discovery Challenge,2008:19.
［6］	Hu M,Liu B.Mining and summarizing customer reviews［C］∥Proc of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining（KDD）,2004:168177.
［7］	Lu Y,Zhai C X,Sundaresan N.Rated aspect summarization of short comments［C］∥Proc of the 18th International Conference on World Wide Web （WWW）,2009:131140.
［8］	Wang H,Lu Y,Zhai C.Latent aspect rating analysis on review text data:a rating regression approach［C］∥Proc of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining （KDD）,2010:783792.
［9］	Wang S G,Yin X Q,Li R,et al.Sentiment clustering of evaluation object based on incomplete information systems［J］.Journal of Chinese Information Processing,2012,26（4）:98102.(in Chinese)
［10］	Zhang M L,Zhou Z H.A review on multilabel learning algorithms［J］.IEEE Transactions on Knowledge and Data Engineering,2014,26（8）:18191837.
［11］	Shen X,Boutell M,Luo J,et al.Mutilabel machine learning and its application to semantic scene classification［C］∥Proc of the 2004 International Symposium on Electronic Imaging,2004:1822.
［12］	Read J,Pfahringer B,Holmes G,et al.Classifier chains for multilabel classification［J］.Machine Learning,2011,85（3）:333359.
［13］	Tsoumakas G,Vlahavas I.Random klabelsets:An ensemble method for multilabel classification［C］∥Proc of the 18th European Conference on Machine Learning （ECML）,2007:406417.
［14］	Zhang M L,Zhou Z H.MLkNN:A lazy learning approach to mutilabel learning［J］.Pattern Recognition,2007,40（7）:20382048.
［15］	Schapire R,Singer Y.BoosTexter:A boostingbased system for text categorization［J］.Machine Learning,2000,39 （2）:135168.
［16］	Elisseeff A,Weston J.A kernel method for multilabelled classification［C］∥Advances in Neural Information Processing Systems,2001:681687.
［17］	Zhang M L,Zhou Z H.Multilabel neural networks with applications to functional genomics and text categorization［J］.IEEE Transactions on Knowledge and Data Engineering,2006,18(10):13381351.
［18］	Spolar N,Cherman E A,et al.A comparison of multilabel feature selection methods using the problem transformation approach［J］.Electronic Notes in Theoretical Computer Science,2013,292（5）:135151.
［19］	Spolar N,Tsoumakas G.Evaluating feature selection methods for multilabel text classification［C］∥Proc of the 1st Workshop on BioMedical Semantic Indexing and Question Answering,2013.
［20］	Zhang M L,JM Pea,V Robles.Feature selection for multilabel Naive Bayes classification［J］.Information Sciences,2009,179（19）:32183229.
［21］	Feng S F,Wang S G.Automobile reviews ontology knowledge base construction oriented towards opinion mining［J］.Computer Applications and Software,2011,28（5）:4547.(in Chinese)
［22］	Bhowmick P K,Basu A,et al.Reader perspective emotion analysis in text through ensemble based multilabel classification framework［J］.Computer and Information Science.2009,2（4）:6474.
	附中文参考文献：
［9］	王素格,尹学倩,李茹,等.基于非完备信息系统的评价对象情感聚类［J］.中文信息学报,2012,26（4）:98102.
［21］	冯淑芳,王素格.面向观点挖掘的汽车本体知识库的构建［J］.计算机应用与软件,2011,28（5）:4547.

[1]	段玲, 郭军军, 余正涛, 相艳, . 基于正文和评论交互注意的微博案件方面识别[J]. 计算机工程与科学, 2022, 44(06): 1097-1104.
[2]	蒋芸，肖潇，侯金泉，陈莉. 基于局部正、负标记相关性的k近邻多标记分类新算法[J]. 计算机工程与科学, 2019, 41(10): 1854-1860.
[3]	蒋芸，肖潇，侯金泉，陈莉. 融合标记独有属性特征的k近邻多标记分类新算法[J]. 计算机工程与科学, 2019, 41(03): 513-519.

基于多标记学习的汽车评论文本多性能识别

Multiple performances identification for car
review texts based on multilabel learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献［21］

相关文章 3

编辑推荐

Metrics

本文评价

基于多标记学习的汽车评论文本多性能识别

Multiple performances identification for car review texts based on multilabel learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 ［21］

相关文章 3

编辑推荐

Metrics

本文评价

Multiple performances identification for car
review texts based on multilabel learning

参考文献［21］