Context injection in expert finding

Abstract

Expert finding is a subject of research in information retrieval and, often, is taken to mean expertise retrieval within a specific organization. The task involves finding an expert on a given topic of interest. Even though there are several proposals in the literature, they do not consider the context in which the given expertise is bound. This paper introduces an approach to inject context into existing expertise evidence based on data extracted from the evidence. Our motivation is to provide context when describing the expertise associated with a candidate expert, allowing a user to understand the results better and choose the best candidate for the task.