| With the rapid development of Internet, IDSS is facing new challenge. While providing more information resources for decision making, it also brings up many uncertainty of information. The picking, quality analysis and data mining become the key problems to solve while using Internet information resources. Internet provides a fast and easy way to communication, make it more easily to select attendants in group decision making. How to set up a system to synthetic the decision opinions is a new problem of group decision making using Internet. This dissertation introduces theory of evidence as the basic theory to IDSS. discusses the frame and decision procedure of Internet oriented and theory of evidence based IDSS, and researches the synthesis of evidences. According to the requirements of application, it synthesis theory of evidence with artificial neural networks. Synergetics, inclusion degree theory, rough sets, researches on the information picking and information quality analysis of Internet information resources, then discusses the aggregation analysis, selection of experts, the synthesis of group experts in group decision making, and the decision with uncertainty based on theory of evidence. The detailed research contains:1) Current states and development trends of IDSS are analyzed. It discusses the affects of Internet to IDSS. analyzes the effect of theory of evidence in the synthesis of experts opinions and the uncertainty decision making, and proposes a framework for the Internet oriented and evidence based IDSS.2) Aims at the information fusion of group decision making, current development of theory of evidence and the current situation of the combination of evidences is discussed. Analyzes the combination method of interrelated evidences and evidences with different importance and with more conflicts, points out the new method and new direction of evidence combination.3) Explores the new method of evidence combination based on ANN and D-S combine rules. It combines the ANN and D-S rules in evidence combination, add learning and data mining capability to evidence combination, make it more suitable for the interrelated, conflicted and with different importance and more suitable for group decision making. Analyze the properties of new combine method, it's advantage contracted with D-S method, and it's filter capability to information resources, then discusses the relation between the combined group decision opinion and the individual expert opinions.4) Specifies an inclusion degree theory based guide line of closeness to illustratethe difference between the uncertainty of expert opinions and standard data, so the decision quality of individual expert can be analyzed. Closeness properties of two uncertainty data is specified, the aggregation analysis method and procedures are designed according to this property. Based on this, proposed a method to select experts based on competition, cooperation, order parameter and server discipline of synergetics, so the forecasting and decision making through group experts opinions combination is more effective.5) Theory of evidence based decision methods are researched. It discusses a decision method through solve the basic believe number based on the focal analysis, relate the resolve of basic believe number with decision process, and relate the decision result with the uncertainty analysis based on the rough set decision rules, rationalize the theory of evidence based decision method.6) Internet oriented and theory of evidence based framework of an IDSS is researched. According to the properties of Internet information, fusing the information fusion method in IDSS. combining qualitative and quantitative analysis, integrate data warehouse, OLAP. data mining technologies, discuss a general framework and decision process to use internet resources.7) Aimed at group expert forecast and investment decision making in the securitiesmarket, an Internet oriented, theory of evidence based securities investment IDSS prototype system is develope...
|
PDF Full Text Request