A study of the relationship between ad hoc retrieval and expert finding in enterprise environment

Abstract

Ad hoc retrieval returns a ranked list of documents in response to a search query, while expert finding returns a ranked list of people in response to an expertise request in the form of a search query, e.g., retrieval. In current state of the art expert finding approaches, ad hoc retrieval is a key component for locating documents relevant to the expertise request. While ad hoc retrieval has been well researched in information retrieval, no previous work has been carried out on the effects of document retrieval in expert finding. The main contribution of this paper is that we are the first to study the effect of document retrieval in expert finding via a background smoothing parameter in a language modeling approach and two document features, namely, anchor text and indegree. Our research gives insight into how to design an effective approach for both ad hoc retrieval and expert finding in enterprise environment. Our experiments on the TREC (Text REtrieval Conference) 2007 Enterprise Track CSIRO (Australian Commonwealth Scientific and Research Organization) dataset shows that background smoothing helps improve ad hoc retrieval but does not help or even hurt expert finding, anchor text helps expert finding but hurt ad hoc retrieval when weighted high, and indegree helps expert finding but does not help improve ad hoc retrieval significantly.