Evidence Theory Based On Rough Set And Its Application In Network Evidence Fusion

With the popularity of the network,the growth rate of cybercrime is very alarming.Therefore,as a new branch of computer security,network forensics has received more and more attention.For such cases,we need to use effective tools and means for network forensics analysis.However,in the process of evidence fusion,we often encounter a wide range of network evidence,excessive network evidence and other issues.Aiming at the complexity of network evidence,this paper proposes a Fuzzy C-means algorithm optimized by reverse genetic algorithm to improve the effect of data preprocessing on network evidence.And this paper presents a fusion method based on variable granularity of rough set,and applies the proposed method to the network evidence fusion.The work of this paper mainly includes these following aspects:(1)This paper studies fuzzy clustering algorithm,and proposes a Fuzzy C-means algorithm based on reverse genetic algorithm.The improved algorithm mainly considered the easy precocious defects of genetic algorithm itself,and worried about that it would lead to a bad effect on the clustering result of improved Fuzzy C-means algorithm.So a reverse genetic algorithm is constructed by introducing reverse learning mechanism into traditional genetic algorithm,and then the reverse genetic algorithm is applied to the improvement of Fuzzy C-means algorithm.Experimental results show that the improved algorithm can effectively improve the clustering accuracy and speed up the iterative efficiency of the whole algorithm.(2)This paper studies the relationship between rough set theory and evidence theory,and proposes a fusion method of evidence theory based on variable granularity rough set.This method deeply analyzes the connection between rough set theory and D-S evidence theory from the perspective of granular computing.Using evidence space distance to judge whether there is conflict between evidences,and constructing a qualitative fusion function based on variable granularity rough set.The example proves that this method can effectively deal with evidence fusion problem in multi-source evidence space.(3)This paper designs a network evidence fusion model based on rough set and evidence theory,and realizes the analysis of network packets.The improved evidence fusion method is applied to the network evidence fusion model.In this system,firstly,the Fuzzy C-means algorithm optimized by reverse genetic algorithm is used to preprocess the collected original data to remove redundancy.And then the evidence fusion method based on variable granularity rough set is used to fuse the evidence,to obtain a consistent and valid evidence fusion result and return the forensics report for the user to view.The results show that the proposed method can effectively improve the effectiveness of network evidence fusion system in evidence fusion.