AVAYA: Sentiment Analysis on Twitter with Self-Training and Polarity Lexicon Expansion

This paper describes the systems submitted by Avaya Labs (AVAYA) to SemEval-2013 Task 2 - Sentiment Analysis in Twitter. For the constrained conditions of both the message polarity classification and contextual polarity disambiguation subtasks, our approach centers on training high-dimensional, linear classifiers with a combination of lexical and syntactic features. The constrained message polarity model is then used to tag nearly half a million unlabeled tweets. These automatically labeled data are used for two purposes: 1) to discover prior polarities of words and 2) to provide additional training examples for self-training. Our systems performed competitively, placing in the top five for all subtasks and data conditions. More importantly, these results show that expanding the polarity lexicon and augmenting the training data with unlabeled tweets can yield improvements in precision and recall in classifying the polarity of non-neutral messages and contexts.