Statistical Machine Translation Models for Personalized Search

Statistical Machine Translation Models for Personalized Search Rohini U ∗ AOL India R& D Bangalore, India Rohini.uppuluri@corp.aol.com Vamshi Ambati Language Technologies Institute Carnegie Mellon University Pittsburgh, USA vamshi@cs.cmu.edu Vasudeva Varma LTRC, IIIT Hyd Hyderabad, India vv@iiit.ac.in Abstract Web search personalization has been well studied in the recent few years. Relevance feedback has been used in various ways to improve relevance of search results. In this paper, we propose a novel usage of rele- vance feedback to effectively model the pro- cess of query formulation and better char- acterize how a user relates his query to the document that he intends to retrieve using a noisy channel model. We model a user profile as the probabilities of translation of query to document in this noisy channel us- ing the relevance feedback obtained from the user. The user profile thus learnt is applied in a re-ranking phase to rescore the search results retrieved using an underlying search engine. We evaluate our approach by con- ducting experiments using relevance feed- back data collected from users using a pop- ular search engine. The results have shown improvement over baseline, proving that our approach can be applied to personalization of web search. The experiments have also resulted in some valuable observations that learning these user profiles using snippets surrounding the results for a query gives bet- ter performance than learning from entire document collection. 1 Introduction Most existing text retrieval systems, including the web search engines, suffer from the problem of “one ∗This work was done when the first and second authors were at IIIT Hyderabad, India. size fits all”: the decision of which documents to re- trieve is made based only on the query posed, with- out consideration of a particular user’s preferences and search context. When a query (e.g. “jaguar”) is ambiguous, the search results are inevitably mixed in content (e.g. containing documents on the jaguar cat and on the jaguar car), which is certainly non- optimal for a given user, who is burdened by having to sift through the mixed results. In order to opti- mize retrieval accuracy, we clearly need to model the user appropriately and personalize search according to each individual user. The major goal of person- alized search is to accurately model a user’s infor- mation need and store it in the user profile and then re-rank the results to suit to the user’s interests using the user profile. However, understanding a user’s in- formation need is, unfortunately, a very difficult task partly because it is difficult to model the search pro- cess which is a cognitive process and partly because it is difficult to characterize a user and his prefer- ences and goals. Indeed, this has been recognized as a major challenge in information retrieval research (et. al, 2003). In order to address the problem of personalization one needs to clearly understand the actual process of search. First the user has an information need that he would like to fulfill. He is the only entity in the process …

Content truncated to 500 words