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

Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 236–241, Denver, Colorado, June 4-5, 2015. c⃝2015 Association for Computational Linguistics ECNU: Using Multiple Sources of CQA-based Information for Answer Selection and YES/NO Response Inference Liang Yi, Jianxiang Wang, Man Lan∗ Shanghai Key Laboratory of Multidimensional Information Processing Department of Computer Science and Technology, East China Normal University, Shanghai 200241, China {51121201035, 51141201062}@ecnu.cn; mlan@cs.ecnu.edu.cn∗ Abstract This paper reports our submissions to com- munity question answering task in SemEval- 2015, which consists of two subtasks: (1) pre- dict 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 answer- s 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 in- formation, answer-specific features, question- specific features, etc. Our submitted primary systems ranked the forth and the second for the two subtasks of English data respectively. 1 Introduction Community Question Answering (CQA) system- s such as Yahoo!Answers rely on users to provide answers (i.e., user generated content) for questions posted. Generally such systems are quite open and the answers provided by users are not always of high quality. For example, a bad answer may present ir- relevant opinions or issues, contain only URL links without direct answer, or even be written informally. Therefore, in order to achieve high-quality user ex- perience and maintain high levels of adherence, it is critical to present high-quality answers and provide direct responses for users. The CQA task in SemEval-2015 (M`arquez et al., 2015) provides such a universal platform for re- searchers to make a comparison between differen- t approaches. This task consists of two subtasks: (1) subtask A is to classify the quality of answers as good, potential or bad, which also refers to the task of answer quality prediction (Jeon et al., 2006; Agichtein et al., 2008); (2) subtask B is to infer the global answer of a YES/NO question to be yes, no or unsure based on individual good answers. Most of the previous research on answer quality prediction has focused on extracting various features to employ ranking or classification methods (Sur- deanu et al., 2011; Shah and Pomerantz, 2010), such as textual features (Agichtein et al., 2008; Blooma et al., 2010) including the length of an answer, over- lapped words between a question-answer (QA) pair, etc. Another kind of widely used feature is extracted from answerer profile information (Shah and Pomer- antz, 2010), such as the number of best answers, the achieved levels and the earned points. However, such information is not often available in real world. Moreover, a recent study (Toba et al., 2014) has tak- en question type into consideration to make the an- swers quality prediction. In this paper, we built two classification systems for the two tasks respectively. For Task A, we ex- …

Content truncated to 500 words