In this paper, we propose a coverage-based genetic knowledge-integration ap
proach to effectively integrate multiple rule sets into a centralized knowl
edge base. The proposed approach consists of two phases: knowledge encoding
and knowledge integration. In the knowledge-encoding phase, each rule in t
he various rule sets that are derived from different sources (such as exper
t knowledge or existing knowledge bases) is first translated and encoded as
a fixed-length bit string. The bit strings combined together thus form an
initial knowledge population. In the knowledge-integration phase, a genetic
algorithm applies genetic operations and credit assignment at each rule-st
ring to generate an optimal or nearly optimal rule set. Experiments on diag
nosing brain tumors were made to compare the accuracy of a rule set generat
ed by the proposed approach with that of the initial rule sets derived from
different groups of experts or induced by various machine learning techniq
ues. Results show that the rule set derived by the proposed approach is mor
e accurate than each initial rule set on its own. (C) 2000 Elsevier Science
Ltd. All rights reserved.