PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Diverse Retrieval via Greedy Optimization of Expected 1-call@k in a Latent Subtopic Relevance Model
Scott Sanner, S Guo, T Graepel, S Kharazmi and S Karimi
In: 20th ACM Conference on Information and Knowledge Management, October 2011, Glasgow UK.

Abstract

It has been previously observed that optimization of the 1-call@$k$ relevance objective (i.e., a set-based objective that is 1 if at least one document is relevant, otherwise 0) empirically correlates with diverse retrieval. In this paper, we proceed one step further and show theoretically that greedily optimizing expected 1-call@$k$ w.r.t.\ a latent subtopic model of binary relevance leads to a diverse retrieval algorithm that shares many features of existing diversification approaches. This new derivation of diverse retrieval is also complementary to a variety of recent research results that have aimed to formally define assumptions that promote diverse retrieval for alternate set-based relevance objectives such as Average Precision and Reciprocal Rank; the derivation presented here for Expected 1-call@$k$ provides a novel perspective on the \emph{emergence of diversity via a latent subtopic model of relevance} -- an idea underlying both ambiguous and faceted subtopic retrieval that have been previously used to motivate the need for diverse retrieval.

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Information Retrieval & Textual Information Access
ID Code:9044
Deposited By:Wray Buntine
Deposited On:21 February 2012