Co-training for Semi-supervised Sentiment Classification Based on Dual-view Bags-of-words Representation

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, pages 1054–1063, Beijing, China, July 26-31, 2015. c⃝2015 Association for Computational Linguistics Co-training for Semi-supervised Sentiment Classification Based on Dual-view Bags-of-words Representation Rui Xia1,2, Cheng Wang1, Xinyu Dai2, and Tao Li3,4 1School of Computer Science, Nanjing University of Science & Technology, China 2State Key Laboratory for Novel Software Technology, Nanjing University, China 3School of Computer Science, Florida International University, USA 4School of Computer Science, Nanjing University of Posts & Telecommunications, China rxia@njust.edu.cn, wangcheng1022@gmail.com, daixinyu@nju.edu.cn, taoli@cs.fiu.edu Abstract A review text is normally represented as a bag-of-words (BOW) in sentiment clas- sification. Such a simplified BOW model has fundamental deficiencies in modeling some complex linguistic phenomena such as negation. In this work, we propose a dual-view co-training algorithm based on dual-view BOW representation for semi- supervised sentiment classification. In dual-view BOW, we automatically con- struct antonymous reviews and model a review text by a pair of bags-of-words with opposite views. We make use of the original and antonymous views in pairs, in the training, bootstrapping and test- ing process, all based on a joint observa- tion of two views. The experimental re- sults demonstrate the advantages of our ap- proach, in meeting the two co-training re- quirements, addressing the negation prob- lem, and enhancing the semi-supervised sentiment classification efficiency. 1 Introduction In the past decade, there has been an explosion of user-generated subjective texts on the Internet in forms of online reviews, blogs and microblogs. With the need of automatically identifying senti- ments and opinions from those online texts, senti- ment classification has attracted much attention in the field of natural language processing. Lots of previous research focused on the task of supervised sentiment classification. However, in some domains, it is hard to obtain a sufficient amount of labeled training data. Manual annota- tion is also very expensive and time-consuming. To address this problem, semi-supervised learning approaches were employed in sentiment classifica- tion, to reduce the need for labeled reviews by tak- ing advantage of unlabeled reviews. The dominating text representation method in both supervised and semi-supervised sentiment classification is known as the bag-of-words (BOW) model, which is difficult to meet the requirements for understanding the review text and dealing with complex linguistic structures such as negation. For example, the BOW representations of two opposite reviews “It works well” and “It doesn’t work well” are considered to be very similar by most statistical learning algorithms. In supervised sentiment classification, many ap- proaches have been proposed in addressing the negation problem (Pang et al., 2002; Na et al., 2004; Polanyi and Zaenen , 2004; Kennedy and Inkpen, 2006; Ikeda et al., 2008; Li et al., 2010b; Orimaye et al., 2012; Xia et al., 2013). Nev- ertheless, in semi-supervised sentiment classifica- tion, most of the current approaches directly ap- ply standard semi-supervised learning algorithms, without paying attention to appropriate representa- tion for review texts. For example, Aue and Ga- mon (2005) applied …

Content truncated to 500 words