Research on trustworthy models for GNN-LSTM-based blockchain cybersecurity enhancement mechanisms

With the increasing complexity of blockchain networks, ensuring data security and trustworthiness remains a significant challenge. This study aims to address these challenges by proposing a trustworthy blockchain cybersecurity enhancement model, integrating Graph Neural Networks (GNN) and Long Short-Term Memory (LSTM) networks. The model enhances transaction behavior analysis and node state prediction, while incorporating Trusted Execution Environments (TEEs) and Hardware Security Modules (HSMs) to establish multi-hop trust paths and dynamic authentication. Experimental results show that the proposed model significantly improves authentication success and stability in high-frequency transaction environments. Overall, the model demonstrates practical potential in securing blockchain networks by offering robust, scalable, and tamper-resistant mechanisms.