K-means Clustering Based User Online Behavior Analysis in P2P Network
Lingfang Wang · Microcomputer applications · 2009
We use the K-means clustering algorithm to categorize users in P2P network from users' online and offline discipline point of view.We find that there exists stable node cluster,as well as day-period node cluster which are complementary to other clusters.We propose that rational and effective use of these user nodes can improve performance and reliability of some P2P systems.