Authors:
Srinivasa Gopi Kumar Peddireddy

Addresses:
Department of Network Implementations and Operations, Charter Communications, Hutto, Texas, United States of America.

Abstract:

The growing complexity and sophistication of cyber-attacks are now gigantic challenges to network security technologies. Those bygone days when reactive, defensive measures could cope with the threats of future security vulnerabilities are gone. A combination of proactive defence technologies and artificial intelligence (AI) is required. A proactive AI-based network defence system is provided in this study that encompasses machine learning techniques, advanced threat detection, and predictive analytics. Automated response, anomaly detection, and real-time data analysis are utilized in the proposed system to identify and remove the threat before exploiting vulnerabilities. The UNSW-NB15 dataset, which was trained and tested on actual network traffic, is utilized by machines such as Random Forest, XGBoost, and Isolation Forest for greater detection efficiency and prevention of false positives. The present work presents better threat detection rates, faster response time, and better resilience than traditional security architecture. Our massive-scale simulated threat attacks-based experiment shows that the AI defence system has a 30% better rate of finding threats and, by 40%, reduces the response time. Our research reveals the necessity of AI to improve network security and its capability further to reduce future cyber-attacks.

Keywords: Artificial Intelligence (AI); Proactive Defense; Network Security; Emerging Threats; Machine Learning; Security Vulnerabilities; Anomaly Detection; Traditional Security Architecture.

Received on: 18/06/2024, Revised on: 05/08/2024, Accepted on: 30/09/2024, Published on: 03/12/2024

DOI: 10.69888/FTSCL.2024.000282

FMDB Transactions on Sustainable Computer Letters, 2024 Vol. 2 No. 4, Pages: 232-241

• Views : 91
• Downloads : 7

Download PDF