Identifying Web Search Query Reformulation using Concept based Matching

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1000–1010, Seattle, Washington, USA, 18-21 October 2013. c⃝2013 Association for Computational Linguistics Identifying Web Search Query Reformulation using Concept based Matching Ahmed Hassan Microsoft Research One Microsoft Way Redmond, WA 98053, USA hassanam@microsoft.com Abstract Web search users frequently modify their queries in hope of receiving better results. This process is referred to as “Query Refor- mulation”. Previous research has mainly fo- cused on proposing query reformulations in the form of suggested queries for users. Some research has studied the problem of predicting whether the current query is a reformulation of the previous query or not. However, this work has been limited to bag-of-words models where the main signals being used are word overlap, character level edit distance and word level edit distance. In this work, we show that relying solely on surface level text sim- ilarity results in many false positives where queries with different intents yet similar top- ics are mistakenly predicted as query reformu- lations. We propose a new representation for Web search queries based on identifying the concepts in queries and show that we can sig- nificantly improve query reformulation perfor- mance using features of query concepts. 1 Introduction Web search is a process of querying, learning and reformulating queries to satisfy certain information needs. When a user submits a search query, the search engine attempts to return the best results to that query. Oftentimes, users modify their search queries in hope of getting better results. Typical search users have low tolerance to viewing lowly ranked search results and they prefer to reformu- late the query rather than wade through result list- ings (Jansen and Spink, 2006). Previous stud- ies have also shown that 37% of search queries are reformulations to previous queries (Jansen et al., 2007) and that 52% of users reformulate their queries (Jansen et al., 2005). Understanding query reformulation behavior and being able to accurately identify reformulation queries have several benefits. One of these benefits is learning from user behavior to better suggest au- tomatic query refinements or query alterations. An- other benefit is using query reformulation predic- tion to identify boundaries between search tasks and hence segmenting user activities into topically co- herent units. Also, if we are able to accurately iden- tify query reformulations, then we will be in a bet- ter position to evaluate the satisfaction of users with query results. For example, search satisfaction is typically evaluated using clickthrough information by assuming that if a user clicks on a result, and possibly dwells for a certain amount of time, then the user is satisfied. Identifying query reformulation can be very useful for finding cases where the users are not satisfied even after a click on a result that may have seemed relevant given its title and sum- mary but then turned out to be not relevant to the user’s information need. Previous work on query reformulation has either focused on automatic query refinement by the search system, …

Content truncated to 500 words