Authors: Mr. Kartik, Dr. Bijendra Singh, Dr. Kavita
Abstract: This research aims to analyze and design an effective ransomware detection technique using machine learning algorithms. The study explores various ML approaches—such as classification, anomaly detection, and clustering—and evaluates their performance in identifying ransomware from normal and benign system behavior. Key features, such as file access patterns, process activities, and network communication, are extracted and analyzed to train and test ML models capable of early detection with high accuracy and low false positives. The primary aims of this study are to understand the behavioral characteristics of ransomware attacks; Identify and select relevant features for effective detection; Evaluate different machine learning models based on precision, recall, F1-score, and accuracy; and Propose a novel or improved ML-based detection framework tailored for real-time ransomware threat identification. This research contributes to the ongoing efforts to fortify cybersecurity by presenting a data-driven, machine learning-powered methodology that enhances early detection capabilities, thereby reducing potential damage and enabling quicker incident response.
DOI:
