The Research On Tacit Knowledge Representation Based On Naturalistic Decision Making

With the fast development of society and the emergence of new high technologies, our life is becoming more and more complicated. Decision-making has been associating with the development and progress of society, such as national decision-making on national economy, enterprise's strategic planning for the next decade, bank's credit risk analysis and so on. The common features of these decisions lie in that the targeted problems are very complex, the decision-making information is not sufficient, the decision-making scenarios are not explicit, and so on. Traditional decision-making theories are of little importance to decision-makers when in face of complicated decision scenarios. In this situation, the decision-makers'professional knowledge and background experience are more reliable. So it is very urgent and also important to transfer organization or expert's tacit knowledge into explicit knowledge by means of calculating and reasoning. In view of the above considerations, the research of tacit knowledge representation based on naturalistic decision-making is made, thus proposing new ideas on how to qualify tacit knowledge into calculable explicit knowledge.The main research work and achievements are as follows:Firstly, the theory of naturalistic decision making is briefly reviewed and the Recognition-Primed Decision (RPD) model is elaborated in details and the relevant applications of naturalistic decision-making are also discussed. Then a literature review on theory and methodology of knowledge representation is made and several different knowledge representation methods are analysed and compared at the same time, including Logic representation, Frame representation, Semantic network representation, Production knowledge representation and so on.Lastly, the structural features of tacit knowledge is analysed and it turns out that Fuzzy Cognitive Map (FCM) can better produce and represent tacit knowledge that has a significant causal relationship.Secondly, based on the previous theoretical research a comprehensive decision making themes and research architecture are presented. Then the structure and causal relationship of Fuzzy Cognitive Map are elaborated and the fundamental procedures and development mathods are also put forward. At the same time, a tacit knowledge representation model based on Fuzzy Cognitive Map is established. Then a new identification method of Fuzzy Cognitive Map is proposed based on Case-based Reasoning (CBR). In the end, a comprehensive study of the reasoning mechanisms and methods of Fuzzy Cognitive Map is made and the reasoning programing and algorithm design of a certain case are also realized.Thirdly, in the end of this paper, a case example analysis of individual credit risk evaluation is maded on the basis of tacit knowledge representation based on Fuzzy Cognitive Map, the recognition and reasoning application analysis of this proposed model are also conducted.The main innovations of this paper lie in that a comprehensive analysis of different naturalistic decision making theory, traditional knowledge represention methods and the main features of tacit knowledge are elaborated and a new casural knowledge representation method based on Fuzzy Cognitive Map is put forward and a calculable tacit knowledge representation model and reasoning method on the basis of naturalistic decision making theory are established at the same time.