Models and Measures for Novel And Diverse Search results (MONADS)

The aim of this project is to research methods for ranking and evaluating automatic information retrieval system results for novelty and diversity. We focus on three related challenges: (1) how to identify different subtopics of a query? (2) how to diversify search results based on the subtopics? (3) how to evaluate diversification strategies?

The name MONADS is meant to reflect the idea that the traditional notion of "relevance" can be decomposed into simpler units---subtopics, aspects, nuggets, or monads of information---and the best way to improve a user's experience with a search engine is to try to identify these smaller units rather than broadly relevant documents.

This material is based upon work supported by the National Science Foundation under grant number IIS-1017026. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.


Project Duration: 9/1/2010 to 8/31/2013



Ben Carterette
Assistant Professor
101 Smith Hall
Dept. of Computer & Information Sciences
University of Delaware
Phone: 302-831-3185

Last updated 11/26/2013.