Research On Information Fusion And Consistency Measure Facing Complex Large Group Of Service Objects

Group decision-making is an important mean to solve various problems of social life. To some extent, the result of the problem made by group can satisfy more people. However, As a result of the rapid development of social economy, an increasing number of more complicated decision problems, like the allocation of half-public service resources which are mostly quantitatively limited have been emerged. Involving large decision group, these complicated problems made the existing decision-making method not applicable any more. So a further discussion about large-group decision problems is needed to solve the practical problems in modern social life.The main idea of this paper is as follows:Chapter one reviews the existing related research and introduces the framework of the paper. The second chapter introduces the concept of large-group decision-making and some existing relevant theory of group decision-making. Besides, it points out some issues that should be probed into. The third chapter studies the clustering method of large group with the following steps:simplify the original data set, determine the clustering center and cluster the updated data set. The fourth chapter puts forward a large-group information fusion method based on reference point, giving the determination of reference point and the weight of each group member. The fifth chapter discussed some issues about consistency measure. It finds out the specific scheme over which some groups may have high-level disagreement as well as the involved groups. In the process of the discovery, two aspects including the differences between groups on the overall scheme and the differences between the schemes reflected in the information given by the whole group are linked. Finally, the last chapter summarizes the main work of the paper and prospect the future research.The contribution of the paper is as following:1. Given the time and space complexity of large-group clustering, the paper gives the specific method for the preprocessing of the original data set, so as to simplify the data set to be clustered.2. The concept of reference point which belongs to the prospect theory is quoted in the process of information fusion and combination reference point is designed based on the existing reference point selection method.3. The paper proposes that the group coordination should give priority to the adjustment of the scheme, rather than the group member’s opinion which reflect their deeply rooted desire and provides the method for finding the most controversial scheme which is potential to be changed.