Collaborative web agent based on friend network

Abstract

Most existing web search tools do not utilize previous search experiences of others to improve performance for the current user. In this article, we present a collaborative web agent designed to enable across-user collaboration in web search and recommendation. We use our distributed collaborative filtering (CF) algorithm based on a P2P overlay network of autonomous agents. Experiments show that our proposed scheme is more scalable than traditional centralized CF filtering systems and alleviates the sparsity problem in distributed CF.