(The) application of knowledge integration using fuzzy logic and genetic algorithms to financial market

Abstract

The literature in the field of decision support shows that decision-making with the integration of knowledge from several sources is more effective than decision-making from a single source of knowledge in solving unstructured problems However, several issues regarding the knowledge integration should be considered to develop an intelligent system with the integration of knowledge. The first issue is to resolve the conflict among knowledge. The integration can provide the enhanced reasoning performance by combining knowledge from multiple sources, however the conflict among information and knowledge supported by these sources arises obviously where a case or an object may be associated with several knowledge creating conflicting situations. This may generate inconsistent knowledge and degrade the performance of an intelligent system when applied to solving an unstructured problem. Therefore, it has to be managed in a suitable way. The second issue is the generation of the optimized reasoning performance. Since the reasoning performance of the integration may be different according to the integration mechanisms, the integration should be achieved appropriately to generate the optimized results. The third issue is the acquisition of problem solving knowledge from human such as experts or users. Numerous studies on bankruptcy prediction have widely applied data mining techniques to finding out the useful knowledge automatically from financial databases, while few studies have proposed qualitative data mining approaches capable of eliciting and representing experts`` problem-solving knowledge from experts`` qualitative decisions. In an actual decision process, the discovery of knowledge from experts is still regarded as an important task because experts`` predictions depend on their subjectivity. The main objective of this research is to develop the mechanisms for the integration of knowledge from multiples sources. The integration mechanisms are based on fuzzy l...