TAPRank: A Time-Aware Author Ranking Method in Heterogeneous Networks

Measuring the impact of authors can not only be a good guidance for new researchers, but also provide a standard for academic foundations and awards. Heterogeneous networks can capture more information about the interactions between entities and they are more and more widely used for the measurement of author impact. However, most of the existing researches take all the papers into the networks as equal, although they have different importance levels. In this paper, we propose a new model: TAPRank, which calculates author impact in author-paper network with considering the PageRank scores of papers for the first time. The PageRank algorithm is implemented in paper citation network, taking the time of publication of each paper into consideration. In addition, the experiments on DBLP dataset show a better performance of TAPRank than other state-of-the-art models.