Exploration and Evaluation of Citation Networks

This paper deals with the definitions, explanations and testing of the PageRank formula modified and adapted for bibliographic networks. Our modifications of PageRank take into account not only the citations but also the coauthorship relationships. We verified the capabilities of the developed algorithms by applying them to the data from the DBLP digital library and subsequently by comparing the resulting ranks of the sixteen winners of the ACM SIGMOD E.F.Codd Innovations Award from the years 1992 till 2007. Such ranking, which is based on both the citation and co-authorship information, gives better and more fair-minded results than the standard PageRank gives. The proposed method is able to reduce the influence of citation loops and gives the opportunity for farther improvements e.g. introducing temporal views into the citations evaluating algorithms.