A hybrid genetic knowledge-integration strategy

• Published in: 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360) pp. 587 - 591
• Main Authors: Ching-Hung Wang, Tzung-Pei Hong, Shian-Shyong Tseng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1998
• Subjects: Genetic algorithms; Laboratories; Neoplasms; Telecommunications
• ISBN: 9780780348691; 0780348699
• DOI: 10.1109/ICEC.1998.700094

Proposes a hybrid genetic knowledge integration approach to effectively integrate multiple rule sets into a centralized knowledge base. The proposed approach consists of two phases: knowledge integration and knowledge refinement. In the knowledge integration phase, rule sets from different sources are integrated to generate good offspring rule sets by an extension of the Pittsburgh approach (S.F. Smith, 1980). In the knowledge refinement phase, the rule sets derived from the knowledge integration phase are then refined to promote their performance by an extension of the Michigan approach (J.H. Holland et al., 1983). Experiments on diagnosis of brain tumors are made to compare the accuracy of the resulting rule set generated by the proposed approach with that of the initial rule sets. Results show that the rule set derived by the proposed approach is much more accurate than each initial rule set.