DiegoLab16 at SemEval-2016 Task 4: Sentiment Analysis in Twitter using Centroids, Clusters, and Sentiment Lexicons

Proceedings of SemEval-2016, pages 209–214, San Diego, California, June 16-17, 2016. c⃝2016 Association for Computational Linguistics DiegoLab16 at SemEval-2016 Task 4: Sentiment Analysis in Twitter using Centroids, Clusters, and Sentiment Lexicons Abeed Sarker and Graciela Gonzalez Department of Biomedical Informatics Arizona State University Scottsdale, AZ 85259, USA {abeed.sarker,graciela.gonzalez}@asu.edu Abstract We present our supervised sentiment classifi- cation system which competed in SemEval- 2016 Task 4: Sentiment Analysis in Twitter. Our system employs a Support Vector Ma- chine (SVM) classifier trained using a num- ber of features including n-grams, synset ex- pansions, various sentiment scores, word clus- ters, and term centroids. Using weighted SVMs, to address the issue of class imbalance, our system obtains positive class F-scores of 0.694 and 0.650, and negative class F-scores of 0.391 and 0.493 over the training and test sets, respectively. 1 Introduction Social media has evolved into a data source that is massive and growing rapidly. One of the most popular micro-blogging social networks, for exam- ple, is Twitter, which has over 645,750,000 users, and grows by an estimated 135,000 users every day, generating 9,100 tweets per second.1 Users tend to use social networks to broadcast the latest events, and also to share personal opinions and experiences. Therefore, social media has become a focal point for data science research, and social media data is be- ing actively used to perform a range of tasks from personalized advertising to public health monitor- ing and surveillance (Sarker et al., 2015a). Because of its importance and promise, social media data 1http://www.statisticbrain.com/ twitter-statistics/ Accessed on: 23rd Decem- ber, 2015. has been the subject of recent large-scale annota- tion projects, and shared tasks have been designed around social media for solving problems in com- plex domains (e.g., Sarker et al. (2016a)) While the benefits of using a resource such as Twitter in- clude large volumes of data and direct access to end- user sentiments, there are several obstacles associ- ated with the use of social media data. These include the use of non-standard terminologies, misspellings, short and ambiguous posts, and data imbalance, to name a few. In this paper, we present a supervised learning approach, using Support Vector Machines (SVMs) for the task of automatic sentiment classification of Twitter posts. Our system participated in the SemEval-2016 task Sentiment Analysis in Twitter, and is an extension of our system for SemEval2015 (Sarker et al., 2015b). The goal of the task was to au- tomatically classify the polarity of a Twitter post into one of three predefined categories— positive, nega- tive and neutral. In our approach, we apply a small set of carefully extracted lexical, semantic, and dis- tributional features. The features are used to train a SVM learner, and the issue of data imbalance is addressed by using distinct weights for each of the three classes. The results of our system are promis- ing, with positive class F-scores of 0.694 and 0.650, and negative class F-scores of 0.391 and 0.493 over the training and test sets, respectively. 2 Related Work Following the …

Content truncated to 500 words