Each expert system provides its users with a great amount of knowledge of a domain. Moreover, a friendly system-user interface usually guarantees an easy access to this knowledge. However, an expert system is sometimes too slow to serve its purpose. Most of expert systems are being implemented as rule-based systems with uncertainty (uncertain knowledge, inexact reasoning). There are a few methods of certainty factors calculation. The well known Dempster's rule of combination always gives a result which is precise and independent of a strategy of conflict resolution. In this paper, we propose a new method, called the rule of convergence. Giving an approximate final result, it can significantly speed up the whole process of reasoning. The methods of uncertain knowledge representation and the methods of inexact reasoning are willingly implemented in medical expert systems. The rule of convergence can really raise their standard of efficiency.
CITATION STYLE
Jankowska, B. (2004). How to speed up reasoning in a system with uncertainty? In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3029, pp. 817–826). Springer Verlag. https://doi.org/10.1007/978-3-540-24677-0_84
Mendeley helps you to discover research relevant for your work.