Expert finding by means of plausible inferences

Expert finding has become an important retrieval task. Expert finding is about finding people rather than documents and the goal is to retrieve a ranked list of candidates/experts with expertise on a given topic. In this paper, we describe an expert- finding system that reasons about the relevance of a candidate to a given expertise area. The system utilizes plausible inferences to infer the relevance of a candidate to a given topic. Experiments are conducted using the TREC 2006 enterprise track text collection. The results indicate the usefulness of our approach.