Detection of Network Intrusions Using Anomaly Detection

Detection of Network Intrusions Using Anomaly Detection

연구 분야: Safety

학회: 2023 20th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA)

The proliferation and widespread use of new technologies has inevitably brought an increase in the occurrence of cyber-attacks. The adoption of cybersecurity technologies and policies are increasingly a central point that requires innovative and appropriate solutions. This paper proposes a solution that aims to help organisations to identify, detect, manage, and respond to security threats assertively and timely. The proposal includes modeling the behavior of network traffic using machine learning and the automatic detection of behavioral deviations that may indicate the occurrence of malicious activity in the network. Two simulations of cyberattacks were conducted, and the results show that is possible to detect anormal network behaviors by means of machine learning analysis. These results are important to improve the visibility of operators and to enable the digital forensic and incident response faster, reducing the potential impacts of cyberattacks.