An Auto-Regulative Document Recommendation System Based on P2P Networks
Feng Guo, Shaozi Li, Kunhui Lin · 2007
Personal recommendation systems help people to find interesting things and they are widely used in the world, such as Amazon.com and eBay.com. This paper presents our recent research work on our auto-regulative personal document recommendation system based on pure P2P network. The peers in our system automatic share and recommend documents with the other peers without central control. The peers can join and leave groups automatic, and the groups can accept and expel peers automatic too, the whole system is auto-regulative. Comparing to the other traditional CIS model systems, ours is easier to extend. In this paper, we present a new algorithm for reputation management and an improvement in the recommendation algorithm, and it works well in the unstructured P2P network and avoids the cold start problem. The results of our experiment show the advantages of the document recommendation system.