ECNU: Using Multiple Sources of CQA-based Information for Answers Selection and YES/NO Response Inference

This paper reports our submissions to community question answering task in SemEval2015, which consists of two subtasks: (1) predict the quality of answers to given question as good, bad, or potentially relevant and (2) identify yes, no or unsure response to a given YES/NO question based on the good answers identified by subtask 1. For both subtasks, we adopted supervised classification method and examined the effects of heterogeneous features generated from community question answering data, such as bag-of-words, string matching, semantic similarity, answerer information, answer-specific features, questionspecific features, etc. Our submitted primary systems ranked the forth and the second for the two subtasks of English data respectively.