Implementation of Dynamic Scanner to Protect the Documents from Ransomware using Machine Learning Algorithms

Abstract

Now-a-days malware analysis and detection is the most needed tool in today’s world. The malware attack is rapidly increasing in all areas especially in corporate sectors. Though there are plenty of tools were available to detect the malwares, the best solution will be the one which uses the machine learning algorithms. By using this, the model can be trained with different algorithm. Each algorithm produces different accuracy rate. From the experimentation, it is found that Random Forest algorithm is chosen as the best algorithm. The datasets that were fed to the model contains different features such as MD5, DLL Characteristics, Size Of Code etc. With this sample data, the model gets trained with the algorithm that has the best accuracy rate. The trained machine learning model is then saved for later use by the most script. The key achievements of this proposed work is to find a solution to detect the malwares before it affects the system by using the best techniques and by giving the high accuracy rate.