Securing Digital Governance: A Deep Learning and Blockchain Framework for Malware Detection in IoT Networks

Abstract

This research uses deep learning and blockchain frameworks to provide a safe platform that promotes digital governance data exchange and interoperability. Use the bonobo optimization algorithm to start a blockchain-based smart city data authentication approach. This paper presents a Blockchain-based malware detection method and framework that uses AI to account for multiple distributed conditions. An upgraded greedy search algorithm and the XGBoost decision tree construct a two-layer extreme gradient boosting (XGBoost) classification model that detects attacks. Three pre-existing XGBoost significance indices were split and merged based on the model's leaf nodes' tree traversal structural features. The augmented greedy search technique retrieved and imported spectral band variables into the XGBoost model's second layer. Bat method was used to optimize XGBoost modeling parameters. The deployed model increased power consumption per device by 13.5%, while Raspberry Pi devices used 0.2 GB and NVIDIA Jetson devices used 0.42 GB. ML models had 93% f1-scores and 95% detection accuracy on both datasets. Our technology detects malware and attacks in Smart Environments efficiently and accurately, as shown by the models.