A reputation system for peer-to-peer networks
NOSSDAV '03 Proceedings of the 13th international workshop on Network and operating systems support for digital audio and video
Learning to form dynamic committees
AAMAS '03 Proceedings of the second international joint conference on Autonomous agents and multiagent systems
| Hi-index | 0.00 |
This paper deals with the problem of peer agent selection in an unstructured P2P recommendation system. The problem is studied in the context of a collaborative P2P bibliographical data management and recommendation system. In this system, each user is assisted with a personal software agent that helps her/him in managing bibliographical data and recommending new bibliographical references that are known by peer agents. One key issue is to define the set of peer agents that can provide the most relevant recommendations. Here, we treat this problem by using CBR methodology. We aim at enhancing the system overall performances by reducing network load (i.e. number of contacted peers, avoiding redundancy) and enhancing the relevance of computed recommendations by reducing the number of noisyrecommendations. The peer selection learning cycle is described in detail. Experimental results are also provided and discussed.