A neural network approach to Web graph processing

Abstract

In this paper, we will provide an overview of some of the more recent developments in web graph processing using the classic Google page rank equation as popularized by Brins and Page [1], and its modifications, to handle page rank and personalized page rank determinations. It is shown that one may progressively modify the linear matrix stochastic equation underlying the Google page rank determinations [1] to one which may contain neural network formulations. Furthermore the capability of these modifications in determining personalized page ranks is demonstrated through a number of examples based on the web repository WT10G. © Springer-Verlag Berlin Heidelberg 2005.