Friday, 1:30–3:00 PM
Chair: Evgeniy Gabrilovich

Towards Natural Question Guided Search

Alexander Kotov, ChengXiang Zhai

Web search is generally motivated by an information need. Since asking well-formulated questions is the fastest and most natural way to obtain information for human beings, almost all queries posed to search engines correspond to some underlying questions, which reflect the user’s information need. Accurate determination of these questions may substantially improve the quality of search results and usability of search interfaces. In this paper, we propose a new framework for question-guided search, in which a retrieval system would automatically generate potentially interesting questions to users based on the search results of a query. Since the answers to such questions are known to exist in the search results, these questions can potentially guide users directly to the answers that they are looking for, eliminating the need to scan the documents in the result list. Moreover, in case of imprecise or ambiguous queries, automatically generated questions can naturally engage users into a feedback cycle to refine their information need and guide them towards their search goals. Implementation of the proposed strategy requires new methods for content indexing, question generation, ranking and feedback, all of which are examined in detail. The proposed strategy has been implemented in a prototype of a question-guided search engine, which has been evaluated by real users on a subset of Wikipedia. Evaluation results show the promise of this new question-guided search strategy.

A General Framework for Exploring Category Information for Question Retrieval in Community Question Answer Archives

Xin Cao, Gao Cong, Bin Cui, Christian Jensen

Community Question Answering (CQA) has emerged as a popular type of service where users ask and answer questions and access historical question-answer pairs. CQA archives contain very large volumes of questions organized into a hierarchy of categories. As an essential function of CQA services, question search in a CQA archive aims to retrieve relevant historical questions and answers to a query question. In this paper, we propose a new approach to exploiting category information of questions to improve the performance of question search, and apply the approach to existing question search models, including one of the state of the art question search models. Experiments conducted on real CQA data demonstrate that the proposed techniques are capable of outperforming baseline methods significantly.

Clustering Query Refinements by User Intent

Eldar Sadikov, Jayant Madhavan, Lu Wang, Alon Halevy

We address the problem of clustering the refinements of a user search query. The clusters computed by our proposed algorithm can be used to improve the selection and placement of the query suggestions proposed by a search engine, and can also serve to summarize the different aspects of information relevant to the original user query. Our algorithm clusters refinements based on their likely underlying user intents by combining document click and session co-occurrence information. At its core, our algorithm operates by performing multiple random walks on a Markov graph that approximates user search behavior. A user study performed on top search engine queries shows that our clusters are rated better than corresponding clusters computed using approaches that use only document click or only sessions co-occurrence information.


Back to full list of papers