Similarity-Based Content Retrieval in Self-Organizing Peer-to-Peer Networks
Takuya Yamaguchi
Graduate School of Science and Engineering, University in Saitama, Japan.
Andrii Zhygmanovskyi
Graduate School of Science and Engineering, University in Saitama, Japan.
Noriko Matsumoto
Graduate School of Science and Engineering, University in Saitama, Japan.
Norihiko Yoshida
Graduate School of Science and Engineering, University in Saitama, Japan.
*Author to whom correspondence should be addressed.
Abstract
This paper presents dynamic reorganization of peer-to-peer networks to make query routing and content retrieval efficient. The reorganization is conducted which use content similarity information. Unlike other related studies using semantic proximity, the method proposed in this paper relies on folksonomy, which gained wide use in figuring out content similarity in various social networks. The proposed method is designed primarily for flooding-based unstructured P2P networks, however it also can be applied to structured P2P networks such as Tapestry. Simulation-based experiments confirm the effectiveness of the proposed method.
Keywords: P2P, network reorganization, content similarity, folksonomy.