一种基于联合深度学习模型的情感分类方法

doi:10.6040/j.issn.1671-9352.1.2016.PC4

山东大学学报（理学版） ›› 2017, Vol. 52 ›› Issue (9): 19-25.doi: 10.6040/j.issn.1671-9352.1.2016.PC4

一种基于联合深度学习模型的情感分类方法

杨艳,徐冰^*,杨沐昀,赵晶晶

哈尔滨工业大学计算机科学与技术学院机器智能与翻译实验室, 黑龙江哈尔滨 150000

收稿日期:2016-11-25 出版日期:2017-09-20 发布日期:2017-09-15
通讯作者: 徐冰(1975— ),女,副教授,博士,研究方向为情感分析.E-mail:hitxb@hit.edu.cn E-mail:yangyan_hit@163.com
作者简介:杨艳(1993— ),女,硕士,研究方向为情感分析.E-mail:yangyan_hit@163.com
基金资助:
国家自然科学基金资助项目(61402134)

An emotional classification method based on joint deep learning model

YANG Yan, XU Bing^*, YANG Mu-yun, ZHAO Jing-jing

Harbin Institute of Technology, School of Computer Science and Technology, Harbin 150000, Heilongjiang, China

Received:2016-11-25 Online:2017-09-20 Published:2017-09-15

摘要/Abstract

摘要： 针对情感分析问题中长句和短句进行情感分类时不同的建模特点,提出了一种基于联合深度学习模型的情感分类方法。该方法融合长短期记忆模型(LSTM)与卷积神经网络(CNN)对影视评论数据进行情感极性判别,该方法采用LSTM模型对上下文进行建模,通过逐词迭代得到上下文的特征向量,采用CNN模型从词向量序列中自动发现特征,并从局部抽取特征后将局部特征整合成全局特征来提高分类效果。所提出的方法在COAE2016评测的任务2的情感极性分类任务中,其系统准确率获得最好结果。

关键词: LSTM, 情感分类, CNN

Abstract: According to the analysis of emotional problems in the modeling of long and short sentences different characteristics to sentiment classification, this paper proposes a classification algorithm based on the model of joint deep learning. Fusion of long short term memory model of the method(LSTM)and convolutional neural network(CNN)on film reviews emotional polarity discrimination, in the method, LSTM model was used to model context, through the word iteration to get feature vector context, and CNN model was used to automatically discover features from the word vector sequence, and integrating local features from the local feature extraction into global features to improve the classification results. The method proposed in this paper is to obtain the best results of the system accuracy in the task of the COAE2016 evaluation task 2.

Key words: sentiment classification, CNN, LSTM

中图分类号:

TP391.1

杨艳,徐冰,杨沐昀,赵晶晶. 一种基于联合深度学习模型的情感分类方法[J]. 山东大学学报（理学版）, 2017, 52(9): 19-25.

YANG Yan, XU Bing, YANG Mu-yun, ZHAO Jing-jing. An emotional classification method based on joint deep learning model[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2017, 52(9): 19-25.

参考文献

[1] YU H, HATZIVASSILOGLOU V. Towards answering opinion questions: separating facts from opinions and identifying the polarity of opinion sentences[J]. Pediatrics, 2005, 116(3):58-59.
[2] KIM S M, HOVYE. Determining the sentiment of opinions[C] // Proceedings of the 20th International Conference on Computational Linguistics. Barcelona:Association for Computational Linguistics, 2004:1367.
[3] TURNEY PD. Thumbs up or thumbs down?: semantic orientation applied to unsupervised classification of reviews[C] //Proceedings of the 40th Annual Meeting on Association for Computational Linguistics. USA: Association for Computational Linguistics, 2002: 417-424.
[4] HASSAN A, RADEVD. Identifying text polarity using randomwalks.[C] //Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics.Uppsala: Association for Computational Linguistics, 2010: 395-403.
[5] PANG B, LEE L, VAITHYANATHAN S. Thumbs up? sentiment classification using machine learning techniques.[C] //Proceedings of the ACL-02 Conference on Empirical Methods in Natural Language Processing-Volume 10. USA: Association for Computational Linguistics, 2002: 79-86.
[6] DAVIDOV D, TSUR O, RAPPOPORT A. Enhanced sentiment learning using twitter hashtags and smileys[C] // Proceedings of the 23rd International Conference on Computational Linguistics: Posters. Uppsala:Association for Computational Linguistics, 2010: 241-249.
[7] WANG Xiaolong, WEI Furu, LIU Xiaohua, et al. Topic sentiment analysis in twitter: a graph-based hashtag sentiment classification approach[C] //Proceedings of the 20th ACM International Conference on Information and knowledge Management. Scottsdale: ACM, 2011: 1031-1040.
[8] GONZALEZ-IBANEZ R, MURESAN S, WACHOLDER N. Identifying sarcasm in twitter: a closer look.[C] // Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: Short Papers-Volume 2. Portland: Association for Computational Linguistics, 2011: 581-586.
[9] 谢丽星, 周明, 孙茂松. 基于层次结构的多策略中文微博情感分析和特征抽取[J]. 中文信息学报, 2012, 26(1): 73-83. XIE Lixing, ZHOU Ming, SUN Maosong. Hierarchical structure based hybrid approach to sentiment analysis of chinese micro blog and its feature extraction[J].Journal of Chinese Information Processing, 2012, 26(1): 73-83.
[10] SASHIHITHLU S, SOMANS S. Complex sentiment analysis using recursive autoencoders[J/OL].Uk: CiteSeerX, 2013.[2015-09-30]. https://core. ac. uk/display/23426251.
[11] SOCHER R, HUVAL B, MANNING C D, et al. Semantic compositionality throughrecursive matrix-vector spaces[C] //Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning.USA:Association for Computational Linguistics, 2012: 1201-1211.
[12] SOCHER R, PERELYGIN A, WUJ Y, et al. Recursive deep models for semantic compositionality over a sentiment treebank[C] //Proceedings of the Conference on Empirical Methods in Natural Language Processing. Seattle:EMNLP, 2013: 1631- 1642.
[13] KIM Y. Convolutional neural networks for sentence classification[J]. ArXiv Preprint, 2014,arXiv:1408.5882.
[14] PORIA S, CAMBRIA E, GELBUKH A. Deep Convolutional neural network textualfeatures and multiple kernel learning for utterance-level multimodal sentimentAnalysis[C] //Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Lisbon: Association for Computational Linguistics, 2015:2539-2544.
[15] LIU Pengfei, QIU Xipeng, CHEN Xinchi, et al. Multi-timescale long short-term memory neural network for modelling sentences and documents[C] //Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Beijing:Association for Computational Linguistics, 2015:2326-2335.
[16] LEE J Y, DERNONCOURT F. Sequential short-text classification with recurrent and convolutional neural networks[J]. ArXivpreprint, 2016, arXiv:1603.03827.
[17] ZEILERM D.Adadelta:anadaptive learning rate method[J]. ArXiv Preprint, 2012, arXiv:1212.5701.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

一种基于联合深度学习模型的情感分类方法

An emotional classification method based on joint deep learning model

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

多维度评价

本文评价

推荐阅读 0

[1]	严倩,王礼敏,李寿山,周国栋. 结合新闻和评论文本的读者情绪分类方法[J]. 山东大学学报（理学版）, 2018, 53(9): 35-39.
[2]	孙建东,顾秀森,李彦,徐蔚然. 基于COAE2016数据集的中文实体关系抽取算法研究[J]. 山东大学学报（理学版）, 2017, 52(9): 7-12.
[3]	陈敬,李寿山,周国栋. 基于双通道LSTM的用户年龄识别方法[J]. 山东大学学报（理学版）, 2017, 52(7): 91-96.
[4]	罗毅, 李利, 谭松波, 程学旗. 基于中文微博语料的情感倾向性分析[J]. 山东大学学报（理学版）, 2014, 49(11): 1-7.