Ransomware Attack Detection and Classification Using Machine Learning

Ransomware has become a serious cybersecurity risk, especially in industries like healthcare wherepatient data integrity and confidentiality are crucial. This study examines current approaches to machinelearning-based ransomware detection and categorisation. For data integrity in encrypted contexts, weinvestigate techniques that integrate cryptographic techniques like MD5 and SHA-256 with algorithms likeSupport Vector Machines (SVM) and Random Forests. In hospital settings, where patient data is kept onencrypted local databases, the study focusses on real-time ransomware detection. We highlight potentialresearch directions for improving ransomware defences while discussing issues including model accuracy,data availability, and encryption overheads.