Machine Learning-Based Approach for Real-Time Detection of Botnet Activities within Wireless Network Infrastructures

Abstract

In recent years, botnet attacks have emerged as one of the most prevalent and sophisticated cybersecurity threats, exploiting network vulnerabilities to compromise system integrity, confidentiality, and availability. Traditional security mechanisms, such as signature-based intrusion detection systems, struggle to keep pace with the dynamic and evolving nature of these threats. This study presents a machine learning-based approach for real-time detection of botnet activities within wireless network infrastructures. Using the Kaggle Malware Traffic Analysis Knowledge Dataset (MTA-KDD’19) and the dataset underwent preprocessing procedures including data cleaning, normalization, transformation, and class balancing using SMOTE. Three machine learning algorithms such as Decision Tree, Random Forest, and Artificial Neural Network (ANN) which were implemented and evaluated based on accuracy, precision, recall, and F1-score where the experimental results revealed that the Random Forest classifier achieved the highest performance accuracy of 99.93%, outperforming the Decision Tree and Neural Network models. The findings demonstrate that Random Forest provides superior generalization and robustness in classifying malicious and benign network traffic. The study concludes that machine learning models, particularly ensemble methods, can significantly enhance proactive threat detection and serve as a foundation for real-time cyber defence systems against botnet attacks.