A Novel method for Reviewer Assignment Problem based on Reviewers' Profile and Fuzzy Graph Connectivity Measure

Reviewer Assignment Problem (RAP) is defined as a problem of assigning the most suitable expert to the proposal. which is a very important task for any research organization In literature many solutions have been proposed for solving RAP but most of them do not deal with the imprecision associated with the problem. In this paper, a novel approach has been proposed for assigning proposals to legitimate experts based on their previous history and domain expertise. For the purpose, we have first extracted the most important words from the author's submitted paper and reviewer's published articles by constructing a fuzzy graph and applying fuzzy graph centrality measures on it. We have then created fuzzy sets for the selected keywords and their weights (for both author's papers and reviewer's publications). Afterwards, WordNet have been used to find the diatance beween the fuzzy sets represented for author's papers reviewers' papers. Finally, fuzzy extension principle is applied on the sets to slect the top 3 experts for reviewing the proposal. Experimentation shows the proposed method provides satisfactory results on SemEVAL-2010 dataset.