Describing Papers and Reviewers' Competences by Taxonomy of Keywords

摘要

This article focuses on the importance of the precise calculation ofsimilarity factors between papers and reviewers for performing a fair andaccurate automatic assignment of reviewers to papers. It suggests that papersand reviewers' competences should be described by taxonomy of keywords so thatthe implied hierarchical structure allows similarity measures to take intoaccount not only the number of exactly matching keywords, but in case ofnon-matching ones to calculate how semantically close they are. The paper alsosuggests a similarity measure derived from the well-known and widely-usedDice's coefficient, but adapted in a way it could be also applied between setswhose elements are semantically related to each other (as concepts in taxonomyare). It allows a non-zero similarity factor to be accurately calculatedbetween a paper and a reviewer even if they do not share any keyword in common.