Analyzing Ransomware attack & improving Detection Rate

The purpose is to explore and analyse various malware detection approaches, as well as to propose a strategy to decrease false positives when separating ransomware attack files from benign (non-malicious) files. The goal of this study is to train a model that can detect any unknown file, that is not in the virustotal database, with high accuracy without any external assistance. Finally, we will cross-check with the Virustotal API for that specific Hash value to see whether it can identify it with a fair Threshold value. The resulting accuracy of 95.62 percent demonstrates the method's efficiency in detecting malware. This paper aims to collect data that will aid in filling research gaps in machine learning-based malware.