A Multimedia Information Retrieval Algorithm in P2P Networks Based on the Classification of Peers

The Multimedia Information Retrieval (MIR) in the P2P networks has been widely studied. In this paper, we propose a new comprehensive similarity function to calculate the similarity of peers in the P2P networks so as to classify these peers. We also apply the relevance feedback in the process of retrieval in order to improve the speed and accuracy of retrieval. In simulation, we compare our algorithm to the traditional method on the basis of the performance of the test which includes four types of thousands of files (text, image, video, and audio). The results show that our algorithm performs better on both speed and accuracy.