International Journal of Scientific Research and Engineering Development

The exponential growth in cyber threats has rendered traditional manual investigation methods ineffective. This paper presents a machine learning-based approach for predicting cyber-attacks by modelling the problem as a multi-class classification task. Using the SDN intrusion dataset comprising 1,188,333 observations with 79 features, we analyze network traffic patterns to classify four attack types: DoS, R2L, U2R, and Probe attacks. A comparative analysis of four supervised machine learning algorithms—SVC, Decision Tree, MLP, and Random Forest—is conducted. Experimental results demonstrate that Random Forest achieves the highest accuracy of 95%, outperforming SVC (75%), Decision Tree (88%), and MLP (91%). The proposed system enables early detection of network intrusions, significantly reducing threat identification time.

Devasri K, Dr.P.Menaka,"Predicting Cyber Attacks Using Machine Learning" International Journal of Scientific Research and Engineering Development, V9(1): Page(1947-1951) Jan-Feb 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.