¹Institute of Chinese Information Processing, Beijing Normal University, 19#, WaiDa St., XinJieKou, HaiDian Area, Beijing, China
²China Patent Information Center, 25#, BeiTaiPingZhuang Road, HaiDian Area, Beijing, China

---

Abstract—To recognize effectively the polarity of Chinese sentences using Chinese appraisers, degree adverbs, and negations, this article presents a new rule-based method. First, the method combines three types Chinese words into the pre-defined rules; it employs the word distances of those rules as constraints; it utilizes the strengths of appraisers and degree adverbs as items of the rules. Then, it utilizes the particle swarm optimization (PSO) to obtain the optimal parameters of the rules, such as the thresholds of constraints and the adjustments of the strengths. Furthermore, it uses Chinese lexicon HowNet as the resources of Chinese sentiment. Experiments show that the method realizes higher precision, recall, and F1 than the baseline of Csupport vector classification (C-SVC) on two real-world applications. Moreover, it obtains the second rank for the task 2 of Chinese opinion analysis and evaluation (COAE) in 2011, indicating the availability of the method.

Index Terms—sentiment analysis, text mining, particle swarm optimization, support vector machine

[PDF]

Cite: Wen Xiong, Yaohong Jin and Zhiying Liu, "Chinese Sentiment Analysis Using Appraiser- Degree-Negation Combinations and PSO," Journal of Computers vol. 9, no. 6, pp. 1410-1417, 2014.