{"37075":{"#nid":"37075","#data":{"type":"event","title":"CSE Seminar: Kevyn Collins-Thompson","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EKevyn Collins-Thompson\u003C\/strong\u003E\u003Cbr \/\u003EResearcher, Microsoft Research (Redmond)\u003C\/p\u003E\u003Cp\u003EFor more information please contact Dr. Guy Lebanon at \u003Ca href=\u0022mailto:lebanon@cc.gatech.edu\u0022\u003Elebanon@cc.gatech.edu\u003C\/a\u003E\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003E\u0022Robust algorithms for information retrieval: effective tradeoffs between risk and reward\u0022\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003ECurrent search engines must typically act under great uncertainty, attempting to interpret the intent of few keywords from the user to match against billions of documents to find a few relevant results. A typical Web search involves a series of operations based on the user\u0027s query, from automatic spelling correction and identifying common word variants to the actual document ranking. Such operations tend to have a risk-reward tradeoff depending on how they \u0022bet\u0022 on different solution hypotheses. In this talk I discuss new theoretical models, algorithms, and evaluation methods for estimating and accounting for uncertainty in retrieval algorithms to achieve effective risk-reward tradeoffs.\u003C\/p\u003E\u003Cp\u003EA prime example of a high-risk, high-reward operation is automatic query reformulation that adds related terms to a query - a process known as query expansion.\u0026nbsp; Query expansion can significantly improve ranking quality on average, but even state-of-the-art methods are highly unreliable and can significantly hurt result quality for some queries, which is one reason for their limited deployment in real-world scenarios. I discuss how casting query expansion as a \u003Cbr \/\u003Econstrained optimization problem over a word graph provides a selective, highly effective modeling framework that reduces the number and magnitude of expansion failures with no loss in the strong average-case gain of the underlying expansion algorithm. I also discuss applications of such optimization frameworks to other problems in information retrieval, such as providing diversity in ranking.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EBio:\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003EKevyn Collins-Thompson is a Researcher in the Context, Learning and User Experience for Search (CLUES) group at Microsoft Research. He completed his PhD (2008) in Computer Science at the Language Technologies Institute, Carnegie Mellon University, where his advisor was Jamie Callan.\u0026nbsp; His research focuses on theoretical models, algorithms, and evaluation methods for effective, reliable information retrieval. Other research interests include user models and personalization, modeling evolutional dynamics of text,\u0026nbsp; educational applications of search technology and machine learning, and understanding how the brain acquires language skills.\u0026nbsp; Kevyn also has more than ten years of industry experience as a software engineer and project manager, responsible for shipping advanced features in Office, Windows, Tablet PC, Encarta, and many other products. \u003Ca href=\u0022http:\/\/research.microsoft.com\/en-us\/um\/people\/kevynct\/\u0022 target=\u0022_blank\u0022\u003EVisit his website here\u003C\/a\u003E.\u003C\/p\u003E\u003Cp\u003E~~~~~~~~~~~~~~\u003C\/p\u003E\u003Cp\u003EPlease join us for a reception preceding the seminar outside Klaus 1324, beginning at 1:30 pm\u003Cbr \/\u003E\u0026nbsp;\u003Cbr \/\u003ETo receive future announcements, please sign up to the cse-seminar email list:\u003C\/p\u003E\u003Cp\u003E\u003Ca href=\u0022https:\/\/mailman.cc.gatech.edu\/mailman\/listinfo\/cse-seminar\u0022 target=\u0022_blank\u0022\u003Ehttps:\/\/mailman.cc.gatech.edu\/mailman\/listinfo\/cse-seminar\u003C\/a\u003E\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Robust algorithms for information retrieval: effective tradeoffs between risk and reward."}],"uid":"27154","created_gmt":"2009-10-01 18:39:59","changed_gmt":"2016-10-08 01:45:55","author":"Louise Russo","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2009-12-04T13:00:00-05:00","event_time_end":"2009-12-04T14:00:00-05:00","event_time_end_last":"2009-12-04T14:00:00-05:00","gmt_time_start":"2009-12-04 18:00:00","gmt_time_end":"2009-12-04 19:00:00","gmt_time_end_last":"2009-12-04 19:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"37041","name":"Computational Science and Engineering"}],"categories":[],"keywords":[{"id":"3498","name":"cse graduate programs"},{"id":"3497","name":"cse seminar"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cdiv\u003E\u003Cp\u003ELometa Mitchell\u003C\/p\u003E\u003C\/div\u003E\n\t\t\t\t\t\u003Cdiv\u003E\u003Cp\u003EPhone:\n\t\t\t\t\t\t404-385-4785\u003C\/p\u003E\u003C\/div\u003E\u003Cdiv\u003E\u003Cp\u003EEmail:\n\t\t\t\t\t\t\u003Cspan\u003E\u003Ca href=\u0022mailto:lometa@cc.gatech.edu\u0022\u003Elometa@cc.gatech.edu\u003C\/a\u003E\u003C\/span\u003E\u003C\/p\u003E\n\t\t\t\t\t\u003C\/div\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}