Séminaires RALI-OLST

Parameterized Concept Weighting For Information Retrieval

Michael Bendersky

Center for Intelligent Information Retrieval, University of Massachusetts, Amherst

Wednesday 14 September 2011 at 11:30 AM

Salle 3195, Pavillon André-Aisenstadt


The majority of the current information retrieval models weight the concepts in the search query (e.g., query terms or phrases) in an unsupervised manner, based solely on the collection statistics. In this talk, I will present a parameterized concept weighting framework that goes beyond the unsupervised estimation of query concept weights. In this framework, the weight of each query concept is automatically learned using a parameterized combination of diverse importance features. In addition, the proposed framework is able to learn importance weights not only for the concepts that are explicitly present in the query, but also for the latent concepts that can be associated with the query through several query expansion techniques. Extensive empirical evaluation demonstrates that the parameterized concept weighting consistently and significantly outperforms a wide range of state-of-the-art retrieval models.


Follow this link to subscribe to future RALI-OLST announcements.
http://rali.iro.umontreal.ca/rali/?q=fr/node/1631

See all the weekly talks for the year:

1991 1992 1993 1994 1995 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024