THE ROLE OF ARTIFICIAL INTELLIGENCE IN AUTONOMOUS CYBER THREAT DETECTION AND PREVENTION FOR BUSINESS SYSTEMS

Md Abul Kalam Azad; K M Zubair; Nurtaz Begum Asha; Akhtaruzzaman Khan; Rakib Hassan Rimon

Authors

Md Abul Kalam Azad Master of Buainess Administation (MBA), University of the Potomac, USA
K M Zubair Master of Science in Computer Science, San Francisco Bay University, USA
Nurtaz Begum Asha Master of Business Administration in Digital & Strategic Marketing, Westcliff University, USA
Akhtaruzzaman Khan Master of Science in Computer Science, San Francisco Bay University, USA
Rakib Hassan Rimon Masters of Science in Business Analytics, Grand Canyon University, USA

Keywords:

Cyber security, Artificial Intelligence, Machine Learning, Threat Detection, Network Traffic Analysis and Intrusion Prevention

Abstract

The skyrocketing growth of digital networks and cloud infrastructures has brought forth a level of difficulty never before encountered in protecting information systems against the changing cyber threats. Traditional security systems, which are mostly reliant on fixed signatures and fixed sets of rules, find it hard to identify advanced attacks that use novel vulnerabilities and evolve in line with defensive countermeasures. To overcome these issues, smart analytical models powered by Artificial Intelligence (AI) have become new revolutionary means of cyber security. These systems adopt machine learning, deep learning, and data-driven automation to detect abnormal behavior, anticipate intrusion trends, and anticipate possible breaches and act against them in real time. The analytical method adopted in this study involves the use of the CICIDS2017 reduced dataset, through a scale that reflects network traffic in the real world, including both legitimate and malicious networks. Data preprocessing and model training were executed with the help of Python to guarantee the data compatibility and ideal features representation. The statistical analysis and aggregation of key indicators were performed with the help of Excel, and the dynamic visualization of the distributions of attacks, flow lengths, and relationships between packets was provided through Tableau to make them easier to interpret. The combined application of these tools allowed exploring the behavior of the network holistically, and the visual and analytical clarity of the results was achieved to comprehend the underlying patterns of cyber-attacks. The findings indicate that AI-driven analytical systems work much more effectively than traditional types of detection mechanisms to detect various types of attacks, such as Denial-of-Service (DoS), Brute Force, and Port Scanning. Further, the study highlights the importance of integrating quantitative analysis with smart visualization to reinforce the situation awareness and preventive threats proactively. The AI-based frameworks are scalable and adaptive to the contemporary cyber security defense systems by automating the detection of anomalies and improving the decision-making reliability. The results of the study help to develop intelligent threat detection techniques and demonstrate the significance of continuous learning models, which can be changed depending on the dynamic nature of cyber threats.