Aim:

            Our study aims to introduce a Blockchain-Enabled Security Framework against Ransomware Attacks using Machine Learning to ensure high protection and prevention.

Abstract:

          RANSOMWARE is a type of malware attack in which the attacker encrypts and locks the victim’s data and crucial files, and then demands money (ransom) in exchange for the data to be unlocked and decrypted. Cybercriminals utilise ransomware as a sort of malware (malicious software). If a ransomware attack happens on a device, it either limits access or encrypts the data (files) of the device. Cybercriminals demand ransom (some amount of money) from their victims in exchange for releasing the data (i.e., to provide the decryption key). A close eye and security software are recommended to protect against ransomware outbreak. Most of the organizations, such as financial institutes and healthcare sectors are targeted by ransomware attacks. Ransomware assaults are among the most frightening types of cyber-attacks, and they are not confined to a specific sector or the countries. Blockchain is a tamper-proof technology, which is more secure, robust and decentralized in nature. Features of blockchain can add more security for detection and mitigation of ransomware more effectively. The application of machine learning algorithms and techniques is to identify and recognize patterns, anomalies, or specific objects within data. In this paper, we propose a new blockchain-enabled security framework using machine learning to detect and defend the ransomware attacks.

Proposing System:

          In the proposed system, we focused on antecedence of ransomware attack as well as aftermath. We also provided a demo attack of Locker Ransomware and Crypto Ransomware to get real-time experience and to aware of it. We will also continuously monitor the system, if any suspicious files is detected it will be removed automatically. We will take backup of the user data to recover it later, in case of ransomware attack so that we don’t need to pay ransom to the attacker. In detection on ransomware we will mainly focuses on the portable executable files which plays a major role in gaining access of the system. We will do the detection of ransomware in Portable Executable files using Honeypot dataset and then the features/signatures are extracted followed by Feature Selection, Correlation and Classification using Machine Learning to detect that the executable file is Legitimate or Infected.

Advantages:

          The continuous monitoring of the system and automatic removal of files, in the event of suspicious activity followed by detection of ransomware and recovery of data make this proposal to stand out than the others.