Intelligent Detection System for Multi-Step Cyber-Attack Based on Machine Learning

Abstract

Cyber-attacks involve stifling processes and activities, conciliating data, or restricting data access by carefully modifying computer systems and networks with malware. There has been a significant increase in these types of attacks over time. Due to the rise in complexity and structure, advanced defensive methods are needed. In the face of growing security threats, traditional methods of identifying cyber-attacks are ineffective. In this paper, the intelligent of intrusion a detection system is suggested. Moreover, the suggested system attempts to evaluate the capability of the k-nearest neighbour's algorithm (KNN) in terms of distinguishing between authentic and tampered data. A reliable dataset named the Multi-Step Cyber-Attack Dataset (MSCAD) is utilized to determine the behavior Among the new sorts of attacks. Moreover, 60% of the dataset was utilized for training the model, and a remaining 40% was used for testing. Evaluation metrics like accuracy, precision, recall, and F1 score are used. Experiments suggest that the proposed system-based KNN could enhance detection performance. Moreover, the suggested approach increases detection accuracy while minimizing false alarms.