Research On Business Intelligent Decision With Integrated Case-based Reasoning System

As long as the developing of information technology, Business Intelligence (BI) is attracting more and more attention. The mostly purpose of BI construction is decision support. The decision maker in the enterprise would like to use the previous case which be accumulated previously to make decision. However, because of the difficulty in constructing the experience knowledge, the previous case can not be utilized effectively in decision support. Aiming at this problem, the article puts forward the Case-based Reasoning (CBR) method to deal with the experience knowledge. About the research of the CBR, there are more attention be paid to the integration of CBR and other methods. The Integrated Case-based Reasoning System has been designed to heighten the competence of problem-solving.This article firstly introduces the conception, function and the work process of BI, investigates the developmental direction and the key technology of Business Intelligent Decision Support System (BIDSS).CBR is pushed out as problems exist in the study of BIDSS. It introduces the relative principle and main direction of research and developing methods, where relativity is deeply investigated. Then, the Integrated Case-based Reasoning method has been discussed and a new integrated reasoning system model which mixed with CBR, RBR and Date Mining technology has been proposed.By using these principles and the model, the article raises a framework of BIDSS with Integrated Case-based Reasoning, and analyses the process of business intelligent decision. For the base property of business intelligent decision factors, the requests of case express and case retrieve for business decision factors are analyzed, and frame express is used to design the case framework and case-base, which realizes the high efficiency for case retrieve through building a integrated retrieve strategy and grouping case property. Then a mixed Similarity Measure method has been proposed.Knowledge Discovery in Case Base (KDC) has been discussed. The data mining technology is used in CBR in order to make the knowledge acquisition more automated and further improves the CBR system. Inductive learning approach of decision rules of unknown cases has been presented to obtain the decision rules of unknown cases. Therefore, by using DM in the reasoning system can improve the competence of knowledge-solving of BIDSS.