Combining Coverage with TMPS for Reviewer Assignment

Abstract

A fundamental aspect of peer review is the as-signment of reviewers. With the help of artificial intelligence, assigning reviewers can save time and effort and even achieve better results. The purpose of this paper is to explore how to assign reviewers to a paper based on matching multiple aspects of expertise. So that the assigned reviewer group covers all the aspects of a paper in a complementary manner, rather than covering the expertise only in the major research field of a paper. We extract research domain sets of the papers by prompt tuning. And calculate the research domain coverage score and TMPS score based on the review candidates and the pending papers. Then, we utilize a greedy round algorithm to establish the assigned reviewer groups for each paper. Finally, the reviewer groups will undergo a discrete check for conflicts of interest to validate the ultimate results. Experiments demonstrate that the proposed method considers the coverage of the research domain adequately. Furthermore, it arranges a proper selection order of reviewers for papers.