Botnet Detection Approach Using Graph-Based Machine Learning

Detecting botnet threats has been an ongoing research endeavor. Machine Learning (ML) techniques have widely used for detection with flow-based features. The prime challenges features are that they high computational overhead and do not fully capture network communication patterns. Recently, graph-based ML witnessed a dramatic increase in attention. In networks, graph data offers insights information about patterns between hosts. this paper, we propose model first considers the significance of before developing generalized detecting botnets based on selected important We explore different feature sets using five filter-based evaluation measures derived from various theories such as consistency, correlation, information. Two heterogeneous datasets, CTU-13 IoT-23, were to evaluate effectiveness proposed several supervised algorithms. Experiment results show reduces training time complexity provides bots rate. Our detects types families exhibits robustness zero-day attacks. Compared state-of-the-art flow-, graph-based, our approach achieves higher precision shows competitive accuracy.