An interest-based peer clustering algorithm using ant paradigm

Abstract

The interest-based clustering is one of promising approaches to achieve low-cost search in peer-to-peer file sharing. It organizes the logical overlay network where peers having similar interests are closely located. In this paper, we propose an interest-based peer clustering algorithm using ant paradigm. Our algorithm is inspired by the ant-based clustering algorithm, which is one of heuristic methods to categorize many data items. We also evaluate this algorithm by simulations. © Springer-Verlag Berlin Heidelberg 2006.