Machine Learning Model for Classification and Detection Ransomware

The paper discusses the critical security issues posed by malicious attacks, malware, and ransomware families to computer systems, data centers, web, and mobile applications across various industries and businesses. The traditional anti-ransomware systems are not effective against sophisticated attacks, and therefore, state-of-the-art techniques, such as traditional and neural network-based architectures, can be utilized to develop innovative ransomware solutions. The paper proposes a feature selection-based framework with different machine learning algorithms, including neural network-based architectures, to classify the security level for ransomware detection and prevention. The authors applied multiple machine learning algorithms, such as Decision Tree, Random Forest, Naïve Bayes, Logistic Regression, and Neural Network-based classifiers, on a selected number of features for ransomware classification. All the experiments were performed on one ransomware dataset to evaluate the proposed framework, and the experimental results show that the Random Forest classifiers outperform other methods in terms of accuracy, F-beta, and precision scores.