A rough set-based expert system for diagnosing information system communication networks

Abstract

In information system communication networks diagnostics, delays in troubleshooting happen when there is incomplete information. Diagnostics is a straightforward process if the information is complete enough to deduce the possible causes but there are cases when the information is insufficient to deduce the possible cause. In certain cases, it is possible to design an expert system algorithm that can create diagnostic rules even if the inputted information is incomplete using rough set theory. The design of this expert system algorithm was done by developing a theorem to help on formulating the data structures. The data structures satisfy the conditions of the theorem. Thus, the expert system can output the correct possible cause even if the inputted symptoms are incomplete. The expert system algorithm created diagnostic rules and the rules are verified giving 100% validity by using empirical testing and the possible causes outputted by the expert system were verified by comparing these with historical data which gave a 90% score. This research developed an expert system algorithm that can handle incomplete information.