R-GCN: a residual-gated recurrent unit convolution network model for anomaly detection in blockchain transactions

• Published in: Multimedia tools and applications 2024-03, Vol.83 (40), p.87527-87551
• Main Authors: Rajmohan, R., Kumar, T. Ananth, Sandhya, S. G., Hu, Yu-Chen
• Format: Article
• Language: English
• Subjects: Anomalies; Artificial intelligence; Blockchain; Computer Communication Networks; Computer Science; Convolution; Data Structures and Information Theory; Deep learning; Machine learning; Multimedia Information Systems; Smart grid; Special Purpose and Application-Based Systems; Unmanned aerial vehicles
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-17942-x

The domain of deep learning has provided an exemplary paradigm for how Artificial Intelligence (AI) can be a disruptive technological paragon through Blockchain Technology (BT). Data experts have recently strived to find the quality of a dataset high enough for machine learning by an AI entity to be effective and efficient. Blockchain technology has become a special, innovative, and fashionable technological development. It also guarantees that the data is reliable and valid through its consensus process. However, new protection creates problems like data anonymity and confidentiality. Deep Learning (DL)-based blockchain data security is needed to deal with the problems mentioned above. This paper proposes an integration of the DL and BT systems, which produces highly reliable performance in enhancing data durability and dissemination. Moreover, a new convolution model called Residual-Gated recurrent unit Convolution Network (R-GCN) is proposed to analyze transactions in a blockchain-based platform using the Stochastic Gradient Boosting (SGB) technique. The proposed framework is implemented in the Ethereum environment using Anaconda and Python packages. Also, an analogy of how these models can be applied in a range of smart technologies, such as the Unmanned Aerial Vehicle (UAV), Smart Grid, healthcare, and green infrastructure, is illustrated.