Paper-to-reviewer assignment, based on expertise degree of reviewers and relevance degree between reviewers and papers

Abstract

Automating the process of paper-to-reviewer assignment is a difficult task to be adequately resolved. Many papers concerning the related topics have been published, but there is still scare of systematic research applications. In most real world conference management, the assignment task is carried out manually by the programme committee, lacking of intelligent assigning rules and efficient matching method. The manual assignment is not only of low efficiency but also does not guarantee to result in the best solution. Given such situation, our paper sets out to analyse the problem of reviewer selection and propose a method for automatically matching papers with reviewers. Our objective is to reduce the loads of both programme committee and reviewers and make the conference-paper assignment task effectual. In this paper, we address this issue of paper-to-reviewer assignment and propose a method to model reviewers, based on the matching degree between reviewers and papers by combining preference-based approach and topic-based approach. We explain the assignment algorithm and show the evaluation results in comparison with Hungarian algorithm.Keywords: Paper-to-reviewer assignment, Hungarian algorithm, matching degree, expertise degree, relevance degree between reviewer and paper.Reference to this paper should be made as follows: Li, X. and Watanabe, T. (2014) 'Paper-to-reviewer assignment, based on expertise degree of reviewers and relevance degree between reviewers and papers', Int.

PRATO: An Automated Taxonomy-Based Reviewer-Proposal Assignment System

“…X. Li and Watanabe (2014) propose a new method to assign papers to reviewers automatically based on preference and topic matching. They conclude that their proposed algorithm can outperform the Hungarian algorithm (Kuhn and Yaw (1955)) as it significantly reduces the time consumed by the comparison process.…”