Formulating effective questions for community-based question answering

Abstract

Community-based Question Answering (CQA) services have become a major venue for people's information seeking on the Web. However, many studies on CQA have focused on the prediction of the best answers for a given question. This paper looks into the formulation of effective questions in the context of CQA. In particular, we looked at effect of contextual factors appended to a basic question on the performance of submitted answers. This study analysed a total of 930 answers returned in response to 266 questions that were formulated by 46 participants. The results show that adding a questionnaire's personal and social attribute to the question helped improve the perceptions of answers both in information seeking questions and opinion seeking questions.