<![CDATA[The expansion of networks increases the likelihood of cyberattacks, particularly those involving Mirai infections. This study aims to design and analyze a machine learning model capable of identifying Mirai infection attack anomalies within network infrastructure using the gotham2025 dataset. This dataset encompasses both normal traffic and various forms of Mirai infection attacks, such as ingress tool transfers, file downloads, Mirai C&C communication, reporting, and Merlin activity. Data imbalance was addressed using balancing methods—specifically the SMOTE technique—to prevent model bias. Algorithms such as Random Forest and Support Vector Machine (SVM) were implemented and evaluated based on accuracy, precision, recall, F1-score, confusion matrix, MCC, Kappa, ROC-AUC. The results demonstrate that Random Forest achieved 100% accuracy, while SVM attained 99.95%. Overall, both algorithms exhibited the most consistent performance in detecting the various forms of Mirai infection attacks. This research contributes to the development of machine learning-based intrusion detection systems aimed at enhancing network security.]]>
Copyrights © 2026
