KLUE: Simple and robust methods for polarity classification

Abstract

This paper describes our approach to the SemEval-2013 task on “Sentiment Analysis in Twitter”. We use simple bag-of-words models, a freely available sentiment dictionary automatically extended with distributionally similar terms, as well as lists of emoticons and internet slang abbreviations in conjunction with fast and robust machine learning algorithms. The resulting system is resource-lean, making it relatively independent of a specific language. Despite its simplicity, the system achieves competitive accuracies of 0.70–0.72 in detecting the sentiment of text messages. We also apply our approach to the task of detecting the contextdependent sentiment of individual words and phrases within a message. 1