Optimalisasi Deteksi Anomali Untuk Pemfilteran Log dan Integrasi Dengan SIEM Menggunakan Machine Learning

Harjanto, Salsabila Amalia
Nurhaliza, Mutiara
Sagala, Jody Hezekiah Tanasa

Madani: Multidisciplinary Scientific Journal
Vol 2, No 7 (2024): Madani, Vol 2. No. 7, 2024

Cybersecurity has become a paramount concern in today's digital age, necessitating robust systems like Security Information and Event Management (SIEM) for effective threat detection through log analysis. Traditional methods often prove inadequate due to static rules prone to false positives. In this study, we propose a Machine Learning-based approach to optimize anomaly detection in Hadoop Distributed File System (HDFS) logs. Evaluating Decision Tree, Naive Bayes, Log Clustering, Support Vector Machine (SVM), and Logistic Regression, Log Clustering emerges with the highest accuracy at 98.19% and the highest recall at 56.05% among the models tested. These findings underscore Log Clustering's efficacy in enhancing cybersecurity in big data environments, particularly in its efficiency for integration with SIEM systems.