KUNLPLab:Sentiment Analysis on Twitter Data

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 391–394, Dublin, Ireland, August 23-24, 2014. KUNLPLab:Sentiment Analysis on Twitter Data Abstract This paper presents the system submitted by KUNLPLab for SemEval-2014 Task9 - Subtask B: Message Polarity on Twitter data. Lexicon features and bag-of-words features are mainly used to represent the datasets. We trained a logistic regression classifier and got an accuracy of 6% in- crease from the baseline feature represen- tation. The effect of pre-processing on the classifier’s accuracy is also discussed in this work. 1 Introduction Microblogging sites has become a common way of reflecting peoples’ opinion. Unlike the regular blogs, the size of a message on a mi- croblogging site is relatively small. The need to automatically detect and summarize the sentiment of messages from users on a given topic or prod- uct has gained the interest of researchers. The sentiment of a message can be negative, positive, or neutral. In the broader sense, automat- ically detecting the polarity of a message would help business firms easily detect customers’ feed- back on their product or services. Which in turn helps them improve their decision making by providing information of user preferences, prod- uct trend, and user categories.(Chew and Eysen- bach, 2010; Salethe and Khandelwal,2011). Sen- timent analysis is also used in other do- mains.(Mandel et al.,2012). Twitter is one of the mostly widely used mi- croblogging web site with over 200 million users send over 400 million tweets daily(September 2013). A peculiar characteristic of a Twitter data are as follow: emoticons are widely used, the maximum length of a tweet is 140 character, some words are abbreviated, or some are elongated by repeating letters of a word multiple times. The organizers of the SemEval-2014 has pro- vided a corpus of tweets and posted a task to au- tomatically detect their respective sentiments. Sub task B of Task 9: Sentiment Analysis on Twitter is describe as follows Task B - Message Polarity Classification “Given a message, classify whether the mes- sage is of positive, negative, or neutral sentiment. For messages conveying both a positive and neg- ative sentiment, whichever is the stronger senti- ment should be chosen.” This paper describes the system submitted by KUNLBLab for participation in SemEval-2014 Task 9 subtask B. Models were trained using the LIBLINEAR classification library (Fan et al., 2008). An accuracy of 66.11% is attained by the classifier by testing on the development set. The remaining of the document is organized as follows: Section 2 presents a brief literature re- view on sentiment analysis on Twitter data. Sec- tion 3 discusses the system developed to solve the above task, characteristics of the dataset, prepress- ing on the dataset, and various feature representa- tion. Section 4 illustrates the evaluation results. Section 5 presents conclusion and remarks. 2 Related Work Sentiment analysis has been studied in Natural Language Processing. Different approaches have been implemented to automatically detect senti- ment on texts (Pang et al., 2002; Pang and Lee, 2004; Wiebe and …

Content truncated to 500 words