A Blockchain-Based Framework for Malware Detection Using Behavioral Artifact Analysis

Abstract

Through the analysis of runtime behaviours such as file system operations, network activity, and API requests, machine learning can be used to discover behavioural artefacts, including those left by malware. Patterns suggestive of malicious activity can then be found using algorithms. Using a range of methods, numerous malware detection strategies have been presented in the past ten years. Most of the time, network-based threats like DDoS attacks are recognised threats to network security. This research proposes novel method in web monitoring based behavioral artifacts analysis for malware detection using blockchain architecture and structure learning method. Input is collected as monitored web data and processed for noise removal and normalization. Then this data has been analysed for detection of malware attacks using reinforcement structure learning based Quantile regression with adversarial neural networks. the output gives classified malware attacks and network monitoring has been carried out using decentralized contract based blockchain encryption model. Simulation results revealed that the proposed method performed with an accuracy of 97 %, an F1-score of 96 %, and a mean average precision of 95 % for web-based behavioral data sets.