Evaluation of Similarity Searching Methods for Music Data in Peer-to-Peer Networks

In this paper, we focus on similarity searching for similar acoustic data over unstructured decentralised P2P networks. Similarity is measured in terms of time warping, which can cope with distortion that is naturally present when query-by-content is performed. We propose a novel framework, which takes advantage of the absence of overhead in unstructured P2P networks and minimises the required traffic for all operations with the use of an intelligent sampling scheme. Within the proposed framework we adapt several existing algorithms for searching in P2P networks. A detailed experimental evaluation which compares the performance of the various similarity searching algorithms demonstrates the efficiency of the proposed scheme.