Integrating the Trend of Research Interest for Reviewer Assignment

Jian Jin, Qian Geng, Qian Zhao, Lixue Zhang

Abstract: Reviewer assignment problem in the research field usually refers to invite experts for comments on the quality of papers, projects, etc. Different factors in conventional approaches are reckoned to choose appropriate reviewers, such as the relevance between reviewer candidates and submissions, the diversity of candidates. However, many studies ignore the temporal changes of reviewer interest and the stability of reviewers' interest trend. Accordingly, in this research, three indispensable aspects are analyzed, including the relevance between reviewer candidates and submissions, the interest trend of candidates as well as the authority of candidates. Next, with extracted aspects, the reviewer assignment is formulated as an integer linear programming problem. Finally, categories of comparative experiments are conducted with two large datasets that are built from WANFANG and ArnetMiner, which shows the availability of the proposed approach in modeling the temporal changes of reviewers' research interest. Also, it demonstrates the effectiveness of the proposed approach for the reviewer assignment problem.