Research On Distance Entropy Based Decision-making Information Fusion Method

Frequent incidents have attacked people in recent years, which cause huge loss of human lives and properties. The development of society and changeable environment have brought great hardship to emergency management. On the one hand, not only the complexity of emergency event, but also the variability nature of environment, make it difficult to avoid loss. On the other hand, the inflation of information result in extracting useful elements from existing ones hardly, thus increasing the information processing cost. The construction of effective emergency management system needs to be further improved. How to make emergency decision rapidly and accurately based on existing knowledge and information is of practical significance. Available researches have made many achievements in emergency decision-making. However, limitations still need to be deeply solved such as massive knowledge management, information acquisition, and collaborative decision-making.According to the deficiency of existing studies, based on the characteristics of emergency decision-making, this paper proposed a decision-making information fusion method from the perspective of cognitive science. In summarize, the paper contains three parts. Firstly, a framework for decision-level information fusion is presented from the perspective of cognitive science, which combines the advantages of information processing like fault tolerance and flexibility, to the typical pattern of emergency decision-making. Moreover, traditional top-down information acquisition mode is replaced by cognition navigated information management. Secondly, concerned with the nature of decision information fusion, knowledge element model in model management is modified into knowledge element model in decision information fusion. Information fusion set is a combination of all information units, who come from the instantiation of knowledge elements. At last, a decision-making information fusion method based on Distance entropy according to the demand for objectivity and scientificity is proposed with an element of distance added into the traditional entropy weight method. Thus we can obtain both local and global fusion weight, finally to gain the global fusion results using the way of linear weighting. Furthermore, the application value and effectiveness of this method is illustrated by empirical analysis.This paper has some exploration meaning to the study on emergency management. Compared with traditional methods, the presented method can achieve better results more effectively and objectively. The advantages are showed as follows. First of all, the decision-level information fusion framework is a good application of decision level information fusion in emergency domain. Information fusion can both handle error and multi-source information problems, which is beneficial for incomplete and unreliable information deficiency in emergency decision-making. Furthermore, obtain and manage information by the navigation of knowledge, can not only reduce redundance, but also cut the cost of data processing, consequently increasing the efficiency of information management. Next, the establishment of knowledge element model in decision information fusion provides effective support for emergency decision making to fuse multi-area and multi-disciplinary knowledge. The recessive description of knowledge element correlation makes it convenient for the self-generation and inference of knowledge element network. This can not only reduce the workload of correlation description, but also enhanced its extensibility. Thus the problem of massive and redundant knowledge base is properly managed. Last but not least, the decision-making information fusion method based on Distance entropy is an incorporation of the ideas of distance into entropy weight. Compared with existing distance method, it plays as a useful tool to estimate the quantity and quality of information objectively. Relatively, it also has absorbed the advantages in consistency measurement in contrast with entropy method, avoiding unnecessary deviation. Analysis shows that the proposed method can solve problems of the hugeness of knowledge base and the objectivity and scientificity of fusion results in decision-making information fusion.