Ransomware Detection Using Aggregated Random Forest Technique with Recent Variants
Julian Rafapa, Arthur Konokix · 2024
The increasing sophistication and frequency of ransomware attacks have posed significant challenges to existing cybersecurity measures, highlighting the need for more effective detection techniques. A novel approach is presented that leverages an aggregated random forest technique to enhance the accuracy and robustness of ransomware detection. Through the integration of multiple random forests, the proposed method demonstrates superior performance in detecting a wide array of ransomware variants, including those utilizing advanced evasion tactics. The methodology includes comprehensive data collection, feature extraction, and rigorous evaluation, yielding high detection accuracy while maintaining low false positive and negative rates. Comparative analysis with other machine learning techniques, such as Support Vector Machines and Neural Networks, further underscores the efficacy of the aggregated random forest model, which also excels in detection speed and resource utilization. The implications for cybersecurity are profound, offering a scalable and efficient solution for real-time ransomware detection, thereby contributing to the resilience of security infrastructures across various sectors. Future work could explore the model's adaptability to new ransomware variants, integration with other machine learning techniques, and broader applicability to different types of malware.