Cyber Situation Awareness Perception Model for Computer Network

Abstract

With the increase in cyber threats, computer network security has raised a lot of issues among various companies. In order to guide against all these threats, a formidable Intrusion Detection System (IDS) is needed. Various Machine Learning (ML) algorithms such as Artificial Neural Network (ANN), Decision Tree (DT), Support Vector Machine (SVM), Naïve Bayes, etc. has been used for threat detection. In light of the novel threats, there is a need to use a combination of tools to accurately enhance intrusion detection in computer networks, this is because intruders are gaining ground in the cyber world and the side effects on organizations cannot be quantified. The aim of this work is to provide an enhanced model for the detection of threats on the computer network. The combination of DT and ANN is proposed to accurately predict threats. With this model, a network administrator will be rest assured to some extent based on the prediction of the model. Two different supervised machine algorithms were hybridized in this research. NSL-KDD dataset was deployed for the simulation process in WEKA environment. The proposed model gave 0.984 precision, 0.982 sensitivity and 0.987 accuracy.

Network Security Situation Prediction Technology Based on Fusion of Knowledge Graph

“…The constantly emerging new technologies have made the situation of network security threats more complex and ever-changing. For traditional network security defense systems, relying solely on security devices is no longer sufficient to cope with constantly evolving network attack methods [1][2]. Therefore, a new security concept-network security situational awareness has emerged.…”

Situation Element Extraction Based on Fuzzy Rough Set and Combination Classifier

“…He and Zhu combined the GRU neural network and particle swarm optimization method to mine the time relationship of data, used the attention mechanism to measure the situation value of current network, and constructed a neural network situation awareness model with the attention mechanism [ 11 ]. Samuel effectively combines a decision tree and artificial neural network to build an enhanced network situation awareness model [ 12 ]. The authors have also done some in-depth research in the field of network security situational awareness and achieved a series of relevant results [ 13 – 17 ].…”