Taxonomy-Driven Expertise Aggregation on Bibliographic Data

Abstract. Finding relevant experts is essential for collaboration. Many Expert Finder applications succeed in finding experts, only at coarse levels of granularity, and within specific domains. Finding experts at fine granularity however, is often needed for effective functioning. Semantics can refine the granularity at which the expertise of various experts can be determined, by explicitly expressing relationships between topics and subtopics in a taxonomy. Such relationships allow extrapolation of expertise, based on the notion that expertise in subtopics is also indicative of expertise in a topic itself. Furthermore, a taxonomy enables enrichment of Expertise Profiles, through distinct relationships between the topics of expertise items and the taxonomy hierarchy. We present semantic techniques for finding experts and expertise at fine granularity, using the implicit coauthorship network of bibliographic data and a taxonomy of topics. Various collaboration levels among researchers, based on degrees of separation, discover unknown researchers with considerable expertise