Automated Question Answering System for Community-Based Questions

摘要

The emergence of community question answering sites, such as, Yahoo! Answer (Y!A), and Quora, indicate that for certain information needs, users prefer receiving focused answers to their questions, rather than a list of URLs from search results. This trend has sparked a rich area of investigation at the intersection of Information Retrieval (IR), Natural Language Processing (NLP), and Machine Learning (ML) of Automated Question Answering (QA). In this paper, we present our attempt at developing an efficient QA system for both factoid and non-factoid questions from any domain. Empirical evaluation of our system using multiple datasets demonstrates that our system outperforms the best system from the TREC LiveQA tracks, while keeping the response time to under less than half a minute.