Recommending collaborators using keywords

This paper studies the problem of recommending collabora-tors in a social network, given a set of keywords. Formally, given a query q, consisting of a researcher s (who is a mem-ber of a social network) and a set of keywords k (e.g., an article name or topic of future work), the collaborator rec-ommendation problem is to return a high-quality ranked list of possible collaborators for s on the topic k. Extensive efiort was expended to define ranking functions that take into consideration a variety of properties, including struc-tural proximity to s, textual relevance to k, and importance. The efiectiveness of our methods have been experimentally proven over two large subsets of the social network deter-mined by DBLP co-authorship data. The results show that the ranking methods developed in this paper work well in practice.