CLaC-SentiPipe: SemEval2015 Subtasks 10 B, E, and Task 11

CLaC Labs participated in two shared tasks for SemEval2015, Task 10 (subtasks B and E) and Task 11. The underlying system configuration is nearly identical and consists of two major components: a large Twitter lexicon compiled from tweets that carry certain selected hashtags (assumed to guarantee a sentiment polarity) and then inducing that same polarity for the words that occur in the tweets. We also use standard sentiment lexica and combine the results. The lexical sentiment features are further differentiated according to some linguistic contexts in which their triggers occur, including bigrams, negation, modality, and dependency triples. We studied feature combinations comprehensively for their interoperability and effectiveness on different datasets using the exhaustive feature combination technique of (Shareghi and Bergler, 2013a; Shareghi and Bergler, 2013b). For Subtask 10B we used a SVM, and a decision tree regressor for Task 11. The resulting systems ranked ninth for Subtask 10B, fourth for Subtask 10E, and first for Task 11.