A language modeling framework for expert finding

Abstract

a b s t r a c tStatistical language models have been successfully applied to many information retrieval tasks, including expert finding: the process of identifying experts given a particular topic. In this paper, we introduce and detail language modeling approaches that integrate the representation, association and search of experts using various textual data sources into a generative probabilistic framework. This provides a simple, intuitive, and extensible theoretical framework to underpin research into expertise search. To demonstrate the flexibility of the framework, two search strategies to find experts are modeled that incorporate different types of evidence extracted from the data, before being extended to also incorporate co-occurrence information. The models proposed are evaluated in the context of enterprise search systems within an intranet environment, where it is reasonable to assume that the list of experts is known, and that data to be mined is publicly accessible. Our experiments show that excellent performance can be achieved by using these models in such environments, and that this theoretical and empirical work paves the way for future principled extensions.

Integration of scientific and social networks

“…The previously proposed approaches for expert finding obtain their estimator of personal expertise by aggregating the relevance scores of the documents related to a person [4] [6]. Most of these methods estimate the expertise score of each person according to the relations and documents existing in a single environment.…”

“…Most of these methods estimate the expertise score of each person according to the relations and documents existing in a single environment. For example, [4,6] and [17] estimate the expertise score of each person based on his related documents collected from the intranet of an organization and the intranet of a university respectively. However, recently proposed models such as [9] and [10] consider heterogeneous sources of information to improve the quality of expert finding results.…”

“…State-of-the-art algorithms for expert finding rank persons based on the content of their associated documents and relations. Although most of the proposed algorithms for expert finding restrict their analysis to the documents and relations exist in a single environment [4][5][6], recent studies [7, ? ] suggest that analysis of personal expertise should not be necessarily undertaken only using the date of one single environment.…”

“…] suggest that analysis of personal expertise should not be necessarily undertaken only using the date of one single environment. For example, while [4][5][6], simply use the information collected from the interant of an organization to rank people, recent approaches [7, ?] also consider information extracted from web pages (i.e.…”

A User-Oriented Model for Expert Finding

“…Building on either candidate or document models, further refinements to estimating the association of a candidate with the topic of expertise have been explored. For example, instead of capturing the associations at the document level, they may be estimated at the paragraph or snippet level [7,20,24]. Other extensions incorporate additional forms of evidence through the use of priors [14], document structure [33], hierarchical, organizational, and topical context and structure [6,23], and Web data [25].…”