Multi-agent Web Information Retrieval: Neural Network Based Approach

The Web is full of information sources. Currently, retrieving useful information on theWeb is a time-consuming process. In this paper, we propose a multi-agent learning approach to information retrieval on the Web, where each agent collaboratively learns its environment from user’s relevance feedback using a neural network mechanism. Our approach makes it possible to discover information sources associated with useful information and then retrieve that information effectively. First, we present a framework of IR agent and its operation for our multi-agent learning approach. Secondly, we define the multi-agent IR system based on our approach and then describe its training procedure for collavorative information retrieval. Finally, we present the experimental results of our approach, comparing them to those obtained by the approach of traditional meta-search service.