WarwickDCS: From Phrase-Based to Target-Specific Sentiment Recognition

We present and evaluate several hybrid systems for sentiment identification for Twitter, both at the phrase and document (tweet) level. Our approach has been to use a novel combination of lexica, traditional NLP and deep learning features. We also analyse techniques based on syntactic parsing and tokenbased association to handle topic specific sentiment in subtask C. Our strategy has been to identify subphrases relevant to the designated topic/target and assign sentiment according to our subtask A classifier. Our submitted subtask A classifier ranked fourth in the SemEval official results while our BASELINE and μPARSE classifiers for subtask C would have ranked second.