Collaborative SVM classification in scale-free peer-to-peer networks

作者：

Highlights：

• Collaborative SVM method in decentralized, hierarchical & skewed P2P networks.

• Performs transitive propagation of models beyond the direct neighbors.

• Considers vast majority of scarcely connected peers.

• Analysis w.r.t music genre classification, centralized methods and data replication.

• Improves the overall accuracy substantially and keeps the communication cost low.

摘要

•Collaborative SVM method in decentralized, hierarchical & skewed P2P networks.•Performs transitive propagation of models beyond the direct neighbors.•Considers vast majority of scarcely connected peers.•Analysis w.r.t music genre classification, centralized methods and data replication.•Improves the overall accuracy substantially and keeps the communication cost low.

论文关键词：Distributed,Classification,SVM,P2P,Skew

论文评审过程：Received 8 March 2016, Revised 30 August 2016, Accepted 4 October 2016, Available online 14 October 2016, Version of Record 26 October 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.10.008