Information Security Risk Assessment Research Based On Optimized Neural Network

This paper studies the relatively mature evaluation criteria and methods in the field of information security,and analyzes the information security situation of the small Internet of things system.The risk assessment index system of the small Internet of things system is constructed by combining the information security characteristics of the Internet of things.We organically combine the latest research results of AHP,information entropy and neural network with the original risk assessment methods.Propose an improved information security risk assessment model based on AHP and an information security risk assessment and prediction model based on improved neural network.And finally the validity of the model is proved by the simulation experiment.The main research contents and innovation points of this paper include:(1)Construction of risk assessment index system for small Internet of things system.On the basis of reference authority evaluation criteria and methods,combine with risk identification and index system determination method of other risk assessment models;construct the information security risk assessment index system for small Internet of things system based on the security environment of Internet of things.And analyze the risk index.(2)Information security risk assessment model based on Improved AHP.By analyzing the shortcomings of AHP in risk assessment,combined with information entropy and interval fuzzy number,we improved it and built the corresponding information security risk assessment model.In the process of determining the weight of the hierarchy,the entropy weight of each expert is calculated by the entropy of the judgment matrix given by the experts.The weight of entropy is adjusted to make the weight of the hierarchy more objective and reasonable.When determining index weight,we take the value interval instead of the specific evaluation value to reduce the subjectivity of the evaluation and the acceptability of the evaluation result.(3)Information security risk assessment and prediction model based on Improved Neural Network.By studying the existing information security risk assessment method combined with neural network,we use the improved cuckoo search algorithm to optimize the BP neural network in order to overcome the defect of slow convergence rate and easy to fall into the local minimum.And finally,we construct the risk assessment and prediction model based on the improved method.(4)Simulation experiment.Based on the established risk assessment index system for the small Internet of things,the information security risk assessment model of improved analytic hierarchy process and the information security risk assessment and prediction model base on Improved Neural Network are used to simulate a small Internet of things system for testing validity of the proposed model.