An Empirical Study for Chinese Sentiment Classification Based on Machine Learning Techniques

Abstract:

As the customer reviews become more and more on the Internet, It would be significant if these reviews are summarized automatically. Sentiment classification aims at predicting the semantic orientation of customer reviews, positive and negative. In this paper, we gave out the framework of sentiment classification, and empirically studied the performance when used different features, term weighting methods and machine learning methods. The experimental results suggest that using binary occurrence to weight the features is more suitable when used Naïve Bayes, but when used the support vector machine, tfidf-c can get the best performance. Besides, we also find that the sentiment terms are not suitable as features when used the approaches based on machine learning methods.

• Periodical: Advanced Materials Research (Volumes 760-762)
• Pages: 2037-2041
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.760-762.2037
• Online since: September 2013
• Authors: Yi Pan, Jun Hua Zou, Shuai Yuan
• Keywords: Machine Learning (ML), Natural Language, Sentiment Classification

References

[1] Andres Montoyo, Patricio Martinez-Barco, Alexandra Balahur. Sentiment analysis: An overview of the current state of the area and envisaged developments. Decision Support Systems, 2012, vol. 53, no. 4, pp.675-679.

DOI: 10.1016/j.dss.2012.05.022

[2] Bo Pang, Lillian Lee, Shivakumar Vaithyanathan. Thumbs up? Sentiment classification using machine learning techniques. In: Proceedings of EMNLP, pp.79-86, (2002).

DOI: 10.3115/1118693.1118704

[3] Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews. In: Proceedings of the 40th annual meeting of the association for computational linguistics, pp.417-424, (2002).

DOI: 10.3115/1073083.1073153

[4] Minqing Hu, Bing Liu. Mining opinion features in customer reviews. In: Proceedings of the 19th national conference on Artifical intelligence, pp.755-760, (2004).

[5] Xiaowen Ding, Bing Liu, Philip S. Yu. A Holistic Lexicon-Based Approach to Opinion Mining. In: Proceedings of the 2008 International Conference on Web Search and Data Mining, pp.231-240.

DOI: 10.1145/1341531.1341561

[6] Popescu AM, Etzioni O. Extracting product features and opinions from reviews. In: Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pp.339-346, (2005).

DOI: 10.3115/1220575.1220618

[7] Zhongwu Zhai, Hua Xu, Bada Kang, Peifa Jia. Exploiting effective features for chinese sentiment classification. Expert systems with applications, 2011, vol. 38, pp.9139-9146.

DOI: 10.1016/j.eswa.2011.01.047