Network traffic analysis and anomaly detection using machine learning algorithms

Author

Samuel, Ashton

Title

Network traffic analysis and anomaly detection using machine learning algorithms

Abstract

The rapid advancement of technology and the exponential growth of internet usage have significantly increased the complexity and volume of network traffic. Traditional methods of network management and security, such as signature-based intrusion detection systems (IDS) and manual traffic monitoring, are no longer sufficient to handle the dynamic and sophisticated nature of modern network environments. This project aims to enhance network traffic analysis and anomaly detection using machine learning algorithms. The study involves the development and evaluation of machine learning models, specifically logistic regression, decision trees, and random forest, to analyze network traffic data and detect anomalies. The research objectives include optimizing network performance, comparing the effectiveness of different algorithms, and providing insights into their practical implementation. Through comprehensive literature review and experimental analysis, this project highlights the challenges and limitations of traditional methods and demonstrates the superiority of machine learning approaches in terms of scalability, accuracy, and adaptability. The results indicate that machine learning algorithms, particularly random forest, offer significant improvements in network anomaly detection, enabling proactive threat mitigation and enhanced network security. This research contributes to the field of network security by providing a robust framework for integrating machine learning techniques into network management systems, ultimately aiming to develop a scalable, reliable, and real-time network anomaly detection system. The findings underscore the importance of continuous innovation in network security methodologies to address evolving cyber threats effectively.

Date

June 2024

Publisher

BUSE

Keywords

Network management and security,

Intrusion detection systems (IDS),

Network traffic analysis,

Machine learning models,

Network security.

Supervisor

Mr. Matombo,