Ransomware Detector Using Artificial Neural Network

This design demonstrates a new approach to detecting ransomware targeted at Microsoft Windows, combining 2 deep literacy neural network classifiers to produce an ensemble, taking lines as input in Microsoft’s standard PE train format, similar as those with a ‘. exe ’ train extension, and returning a vaticination of the train belonging to 1 of 3 classes benign, general malware, or ransomware. The model’s capability to distinguish between ransomware and other forms of malware allows it to be applied as an extension to a being malware discovery system similar as anti-virus software, and aid in the categorization and rear engineering of new in- the-wild ransomware samples. The results of testing the ensemble model on data not seen in its training suggest a high position of prophetic power in classifying new in- the-wild samples.