Network Security Based on D-S Evidence Theory Optimizing CS-BP Neural Network Situation Assessment









Abstract

The problem of network security is endless. Network security situation assessment becomes a hot research issue. These hotspot problems set in a trusted measure environment, this paper presents a D-S evidence theory situation assessment model based on optimized CS-BP neural network. This model enhances the local searching ability of cuckoo algorithm by conjugate gradient calculation, and introduces it into BP neural network to improve training convergence speed and overcome local minimum problem. When traditional D-S evidence theory is used to evaluate complex network situation, the reliance on expert experience assignment leads to the problem of low accuracy of evaluation. The optimization of CS-BP neural network is chosen to reduce the subjectivity of BPA and improve the accuracy of evaluation. The validity of the model is verified by setting up a test environment.


Modules


Algorithms


Software And Hardware

• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL