Efficient Music Genre Retrieval Based on Peer Interest Clustering in P2P Networks

DOI

10.2991/iske.2007.63How to use a DOI?

Keywords

Music genre classification; Peer interest clustering; Content-based music retrieval; Peer-to-Peer

Abstract

Content-based music retrieval is desirable in Peer-to-Peer (P2P) networks, considering its popularity for users and its ability of semantic search, intensive computing cost raises a barrier to efficiency and scalability though. In this paper, we propose an approach of music genre retrieval based on peer interest clustering. Automatic music feature extraction and adaptive shared music file clustering are described. Peers with similar music genre are clustered together, based on which search mechanism and improvement alternatives are deployed. The results of experiments prove the algorithm increases search performance, including precision and recall while reducing network traffic and peer workload.

Copyright

© 2007, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xianfu Meng
AU  - Zhenyu Guo
AU  - Yue Gong
PY  - 2007/10
DA  - 2007/10
TI  - Efficient Music Genre Retrieval Based on Peer Interest Clustering in P2P Networks
BT  - Proceedings of the 2007 International Conference on Intelligent Systems and Knowledge Engineering (ISKE 2007)
PB  - Atlantis Press
SP  - 373
EP  - 380
SN  - 1951-6851
UR  - https://doi.org/10.2991/iske.2007.63
DO  - 10.2991/iske.2007.63
ID  - Meng2007/10
ER  -