Sentiment Analysis in Twitter at SemEval-2014
Our system expands tokens in a tweet with semantically similar expressions using a large distributional thesaurus and calculates the semantic relatedness of the expanded tweets to word lists representing positive and negative sentiment. This robust approach helps to assess the polarity of tweets that do not directly contain polarity cues. Moreover, we incorporate syntactic, lexical and surface sentiment features. The details are described in our paper.
According to the official results,our system (ukp-dipf) achieved the 8th place on tweet level (subtask B) in terms of macro-averaged F-score among 50 systems, with particularly good performance on the Life-Journal corpus (F 1 =71.92) and the Twitter sarcasm (F 1 =54.59) dataset. On the expression level (subtask A), our system ranked 14 out of 27 systems, based on macro-averaged F-score.