Topic-Sensitive Link-Ranking Approach for Academic Expert Recruiting

The problem of academic expert recruiting is concerned with finding the experts on a specified research field. It has many real-world applications and has recently attracted much attention. However, the existing methods are not versatile and entirely suit for the special requirements from academic area where the co-authorship and the citation relation play important roles in judging researcherspsila achievement. In this paper, we propose and develop a flexible data schema and a topic-sensitive co-pagerank algorithm for studying this problem. The main idea is measuring the authorspsila authorities with considering topics bias on the basis of their social networks and citation networks, and then, recommending expert candidates for the requests. To infer association between authors and topics, we derive a probability model on the basis of latent Dirichlet allocation (LDA) model. We further propose several techniques such as reasoning the interested topics of query, modeling author profile on the supporting documents to instruct the practices. Our experiments show that the proposed strategies are all effective to improve retrieval accuracy.