Ransomware Clustering and Classification using Similarity Matrix

Ransomwares are amongst the most dangerous malwares that face and affect any business by restricting data access or leaking sensitive information over the internet. Ransomwares binary analysis can provide a way to define the relationships between distinct features employed by ransomware families. Malware classification and clustering systems offer an effective malware indexing with search functionalities, similarity checking, samples classification and clustering. Most studies focus on the static and dynamic features extraction, machine and deep learning or visualization techniques used to minimize the false positive detections. However, there are gaps in malware tracking and classification. In this paper, we focus on the static features extraction and calculate the absolute Jaccard Index. The main objective is to identify the suitable static feature to decrease the ransomware sample detection time and the accuracy of ransomware's family tracking and clustering using similarity matrix.