LEARNING BY EXPERIENCES IN A WEB-BASED RULES-BASED CONSULTATION SYSTEM AS A POSSIBILITY TO IMPROVE ITS PERFORMANCE

摘要

Starting time of an application gives users a first impression about the application. It should be as short as possible. A Web-based application will be started when a user calls its Web page with an Internet browser. With Web-based consultation systems, which use the rules-based expert system technology, in the starting time, necessary rules will be loaded into system memory, and be organized into a reasoning network that will be used by the reasoning engine. This task costs much time, especially when a system has got so many rules for reasoning. So all Web-based consultation systems, which use the rules-based expert system technology, have to cope with a challenge of how to reduce the time for building the reasoning network from rules. In this paper, we would like to present a general model of learning from experiences and then apply it for our web-based rules-base consultation system as a solution to solve the problem with the long starting time.