CrowdQ: Crowdsourced Query Understanding

Work in hybrid human-machine query processing has thus far focused on the data: gathering, cleaning, and sorting. In this paper, we address a missed opportunity to use crowdsourcing to understand the query itself. We propose a novel hybrid human-machine approach that leverages the crowd to gain knowledge of query structure and entity relationships. The proposed system exploits a combination of query log mining, natural language processing (NLP), and crowdsourcing to generate query templates that can be used to answer whole classes of different questions rather than focusing on just a specific question and answer.