INF-UFRGS-OPINION-MINING at SemEval-2016 Task 6: Automatic Generation of a Training Corpus for Unsupervised Identification of Stance in Tweets

This paper describe a weakly supervised solution for detecting stance in tweets, submitted to the SemEval 2016 Stance Task. Our approach is based on the premise that stance can be exposed as positive or negative opinions, although not necessarily about the stance target itself. Our system receives as input ngrams representing opinion targets and common terms used to denote stance (e.g. hashtags), and use these features, together with the sentiment detection solutions, to automatically compose a large training corpus. Then, it applies a supervised learning algorithm to develop a stance prediction model.