Towards Capturing Population-Wide Expertise via Online Professional Social Network Systems

Abstract

While much research work has been carried out in relation to the expertise of individuals in the domains of Q&A portals, special interest group portals, Wikipedia, blogs, and conference reviews, very little research has been done in relation to expertise measurement, based upon online professional social networks such as LinkedIn. Existing approaches have been based on evidence-based analysis from textual sources and communications. In this paper, we propose an expertise model based on expertise vectors composed of attributes and values derived by integrating information obtained in the two following ways: semi-automatic calculation utilizing online professional profile information collected from diverse online sources such as professional social network sites and some level of manual expertise identification by individuals. Unlike past approaches focusing on expertise measurement based on analysis of secondary evidence such as textual communications and/or social network activities, this is a novel approach to identifying expertise based on real-world work experience, qualifications and other skills indicated via information found in online professional profiles, often from professional social network sites. Given contemporary professional social network sites can have 100s of millions of participants such automated expertise measurement approaches are a step towards fine-grained, up-to-date and population-wide expertise capture and quantification.