Analysis of the Grid Agent Method in Information Retrieval Field

摘要

Information retrieval has changed considerably in recent years with the expansion of the World Wide Web (WWW). With powerful workstations and parallel processing technology, efficiency is not a bottleneck. But still many problems exist in information retrieval (IR), the most important ones are how to understand the user's needs and how to process user queries with high performance. This article first hopes to explain the model, which includes three kinds of agents: Grid service agent, service request agent and Grid space manager. The main actions involved in Grid information service discovery process are registry, discovery, and access. Through these actions, the Grid space manager will find out the suitable services. Secondly, to implement the adaptivity of the Grid agent, we assume that every Grid agent should remember the success and quality of the former cases by its repository and, so that the agent can gain experience of user's preference. The method which how the Grid agent is able to seek for the information automatically and precisely by its repository and the method of AP (Artificial Psychology) is also emphasized. At last, we conclude highlighting some directions of future research, which are needed in order to understand the formal characteristics of Information Retrieval better.