Y!Q: Contextual Search at the Point of Inspiration Reiner Kraft Yahoo!, Inc. 701 First Avenue Sunnyvale, CA 94089 reiner@yahoo-inc.com Farzin Maghoul Yahoo!, Inc. 701 First Avenue Sunnyvale, CA 94089 fmaghoul@yahoo- inc.com Chi Chao Chang Yahoo!, Inc. 701 First Avenue Sunnyvale, CA 94089 chichao@yahoo-inc.com ABSTRACT Contextual search tries to better capture a user’s informa- tion need by augmenting the user’s query with contextual information extracted from the search context (for example, terms from the web page the user is currently reading or a file the user is currently editing). This paper presents Y!Q—a first of its kind large-scale contextual search system—and provides an overview of its system design and architecture. Y!Q solves two major prob- lems. First, how to capture high quality search context. Second, how to use that context in a way to improve the relevancy of search queries. To address the first problem, Y!Q introduces an informa- tion widget that captures precise search context and pro- vides convenient access to its functionality at the point of inspiration. For example, Y!Q can be easily embedded into web pages using a web API, or it can be integrated into a web browser toolbar. This paper provides an overview of Y!Q’s user interaction design, highlighting its novel aspects for capturing high quality search context. To address the second problem, Y!Q uses a semantic net- work for analyzing search context, possibly resolving am- biguous terms, and generating a contextual digest compris- ing its key concepts. This digest is passed through a query planner and rewriting framework for augmenting a user’s search query with relevant context terms to improve the overall search relevancy and experience. We show experi- mental results comparing contextual Y!Q search results side- by-side with regular Yahoo! web search results. This evalu- ation suggests that Y!Q results are considered significantly more relevant. The paper also identifies interesting research problems and argues that contextual search may represent the next major step in the evolution of web search engines. Categories and Subject Descriptors H.3.3 [Information Systems]: Information Storage and Retrieval—Information Search and Retrieval Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. CIKM’05, October 31–November 5, 2005, Bremen, Germany. Copyright 2005 ACM 1-59593-140-6/05/0010 ...$5.00. General Terms Experimentation Keywords Contextual Search, Web Information Retrieval, Content Anal- ysis, Semantic Networks, Search Personalization, Context 1. INTRODUCTION Today’s web search engines accept keyword-based queries and return results that are relevant to these queries. These engines have proven to be extremely useful, perhaps surpris- ingly so given the short length of a typical web query [19] and the huge size of today’s web corpora. However, rele- vancy is a significant challenge for search engines, especially for queries with an ambiguous topic or intent (such as the popular “jaguar” example.) Contextual web search aims to improve the relevancy of web search results by considering the information context of the user’s current task along with the user’s query. Often, search queries are formulated while the user is en- gaged in some larger task. In these cases, there is often an information context available that can help refine the meaning of the user’s query. For instance, a user may be browsing a web page about the jaguar car. The article stim- ulates some interest (at the “point of inspiration”) and the user then wants to know something related to that car. A contextual search engine might take that web page as an ad- ditional input to disambiguate and otherwise augment the user’s explicit query. Although contextual search has been identified to be a promising direction for improving web search in the liter- ature (e.g., [10], [11], [14]), as of early 2005 there were no major web search engines that would provide a contextual search application for their users. This changed in February of 2005, when Yahoo! launched Y!Q Contextual Search in beta 1 . In this paper we present Y!Q—a first of its kind large-scale contextual search application integrated with a major web search engine (Yahoo! Search 2 .) We provide an overview of Y!Q’s system design and architecture, and point out in- teresting research problems related to different aspects of contextual search. Y!Q evolves around two mechanisms. The first is a novel user interface and interaction model for obtaining high qual- 1 http://yq.search.yahoo.com 2 http://search.yahoo.com