SDEC: A P2P Semantic Distance Embedding Based on Virtual Coordinate System

摘要: Large-scale P2P applications can benefit from the ability to predict semantic distances other peers without having contact them first. In this paper, we propose a novel distance embedding approach, SDEC, in network, which assigns synthetic coordinates such that between of two approximately predicts any peers. Specifically, is quantitatively characterized through vector space model based on peers' profiles, and then, measured peer handful current those peers, adopt spring relaxation method, mimicking physical mass-spring system, simulate procedure, find minimal energy configuration corresponding relatively accurate embedding. Simulation results show 3-dimensional Euclidean embed these with high accuracy.