MACC'97 Session: Information Retrieval
Search engines for World Wide Web have the following problems: (1) it is difficult for most of users to select the proper information resource from among lots of the results which they produce, (2) they often cause the high CPU-load of the server machines and the increase of the network traffic, and (3) they frequently return useless results that are out-dated. In order to solve such problems, we propose a resource discovery mechanism based on the word-of-mouth agent system. The system consists of a number of agents who communicate each other by word-of-mouth. The communication is restricted to a relatively small area and accompanied by evaluation information. Therefore the system is effective to reducing the network traffic and keeping information updated.
Resource Discovery, Information Transmission, Word-of-Mouth, World Wide Web, Search Engine, Evaluation, Social Recommendation
PS
file(+gzip) (in Japanese)