Network security situation evaluation based on modified D-S evidence theory

Abstract

With the rapid development of global information and the increasing dependence on network for people, network security problems are becoming more and more serious. By analyzing the existing security assessment methods, we propose a network security situation evaluation system based on modified D − S evidence theory is proposed. Firstly, we give a modified D − S evidence theory to improve the reliability and rationality of the fusion result and apply the theory to correlation analysis. Secondly, the attack successful support is accurately calculated by matching internal factors with external threats. Multi-module evaluation is established to comprehensively evaluate the situation of network security. Finally we use an example of actual network datasets to validate the network security situation evaluation system. The simulation result shows that the system can not only reduce the rate of false positives and false alarms, but also effectively help analysts comprehensively to understand the situation of network security.With the application of computer network being more widely used and network attack technology constantly renovated, security problems are becoming more and more serious. Although a variety of sensors with different functions are deployed in detection and prevention of attacks, these security tools often generate huge reports with high false positive rate and high false negative rate [1] . The network security situation awareness (NSSA) technology emerge as the times require, it provides a macroscopic view of the current network security, which make the administrator grasp the current network security situation value more quickly and accurately, based on security events and as a basis for taking corresponding measures [2] . Up to date, research on NSSA has not yet formed a unified standard about the NSSA implication and presentation methods [3] . Thus, the network security situation assessment model and its key technology has become the research focus in the field of network security [4] . In order to accurately percept the network security, we need to deploy various types of sensors in network environment. By collecting, analyzing and processing the information from the sensors, we could identify the attack features, such as the source, size, speed, etc [5] . An effective method on NSSA is highly required to help us fuse all available information properly and comprehend the situation value with ease [6] .Bass [7] first put forward the combination of network security and situational awareness. He gave a network security situation awareness framework based on data fusion. The framework was divided into 5 layers to evaluate the security situation by analyzing the identified

(Retracted) Research on computer network security evaluation based on image recognition and neural network

“…It is an acyclic graphical pattern describing the probabilistic relationships of the dependency among stochastic covariates and is the essential graph used in causal analysis 12 . It offers us a clear and succinct graphical model to illustrate interesting relationships between elements and allows for more clarity and easier understanding of intricate associations between elements 13 . The Bayesian network has many nodes and directed maps whose known target node’s location can be determined by a given solution 14 .…”

“…12 It offers us a clear and succinct graphical model to illustrate interesting relationships between elements and allows for more clarity and easier understanding of intricate associations between elements. 13 The Bayesian network has many nodes and directed maps whose known target node's location can be determined by a given solution. 14 A Bayesian network links the introduction of pie charts to variables, so the logical relationships between variables can be easily represented.…”

A New Preprocessing Method on Evidence Weight of Dempster Fusion Rule

“…Defects of Dempster fusion rule. As a method of uncertain knowledge reasoning, D-S evidence theory has been widely applied on uncertain information fusing and reasoning [4] [5] . However, there are some limitations in the application of the conventional D-S evidence theory.…”

Molecular dynamics simulation‐guided toehold mediated strand displacement probe for single‐nucleotide variants detection

“…Border mismatches render high discrimination factor (DF) with a median of 124 which is superior than currently reported results. [17] Probe-sink system [18] was employed to further enhance DF allowing for detecting VAF down to 0.1%. We successfully used the probes to detect SNVs of SARS-CoV-2 and cancer genes with high confidence in plasmid clones (Scheme 1).…”