A Multi-Agent Based Information Locating Strategy with Rewards Evaluation

Abstract

As the application of the large scale information network, the mount of the content in the network is increased sharply. This brings the challenge to the technology of searching and monitoring information content. In this article, it is introduced an information locating strategy, which is based on multi-agent and can be distributed and work parallel. The concept of search-degree and search-rewards are introduced for the agents in the overall system can work simultaneously. The object of work is to search the numerous information nodes or Websites at a very fast speed and not needing occupying too much of network resources. At the same time, some most important information can be monitored for content management. The description and algorithm of the multi-agent information searching system structure and the strategy is given in this paper