Case-based Rule-based reasoning Expert system Technology selection Experience-oriented problems Self-learning
EVANS, L., LOHSE, N. and WEBB, P., 2010. A self-learning case and rule-based reasoning algorithm for intelligent technology evaluation and selection. Presented at the AI-2010 Thirtieth SGAI International Conference on Artificial Intelligence, Cambridge, UK, 14th-16th December.
This research programme proposes to fulfill the existing gap in knowledge by providing an experience-oriented decision algorithm to solve technology selection problems based on cases and expert’s experience. The approach adopts historical case-based data to extract rules through the ID3 rule induction algorithm. The decision model integrates a rule induction approach in a rule-based knowledge system and database management system to support automated knowledge mining and usage. The adoption of a pair-wise comparison algorithm within the similarity index assists in relating the importance of the criteria within the knowledgebases reasoner. A series of historical and new solutions are presented in a scoring index based on the requirements of a new case.