大学サイト   学部サイト   学科サイト   EnglishPages
 
旧尾家・塚本研究室です.尾家祐二 現・九州工業大学学長の情報は こちら.
   
研究業績


Evolutionary P2P Networking for Enhancing Search Performance
著者
  Kei Ohnishi, Kento Ichikawa, Yuji Oie
 
雑誌名/会議名
  Joint 4th International Conference on Soft Computing and Intelligent Systems and 9th International Symposium on advanced Intelligent Systems (SCIS&ISIS 2008), CD-ROM 4 pages, Nagoya, Japan 2008 年 9 月
 
アブストラクト
  The present paper proposes a concept of evolutionary peer-to-peer (P2P) networking in which topologies of a running P2P network are dynamically and adaptively modified by an evolutionary algorithm and an algorithm based on the proposed concept. In addition, we evaluate the proposed concept through simulations. Evolutionary P2P networking allows every node to simultaneously belong to several topologies of a running P2P network. Each node assigns a fitness value to each of the P2P network topologies according to the result of using each network topology. The set of P2P network topologies are reconstructed by applying evolutionary operators to individuals encoding them in a server. In the simulation, nodes are search objects and the number of hops required for finding requested nodes on a certain network topology is the fitness value of the network topology. The simulation results show that the proposed algorithm for evolutionary P2P networking can generate a network topology that provides high search performance for most nodes when object nodes for search are strongly biased.
 
キーワード
  その他
 
記述言語
  English
 
 

web_admin [ at ] infonet.cse.kyutech.ac.jp /  Copyright (c) 2016 Oie & Tsukamoto Laboratory, All Right Reserved.