Smart Intelligence Powered Cyber-Attack Detection in Financial Sector Transactions using Enhanced Proactive Block Chain Security

Cyberattacks in the financial sector are on the rise as strategic and threatening attacks intensify during critical periods in the global economy. Traditional approaches to network operations as a tool for analyzing communication logs are associated with several complexities, insert malicious data into the logs. The inability to detect impacts and low detection accuracy lower recall accuracy and classification accuracy, resulting in security failures and data breaches. To address this issue, we proposed a Hyperactive Deep Belief Generative Adversarial Neural Network (HDB-GANN) algorithm to identify optimal features and increase the accuracy for improved classification. Furthermore, the Z-Score min-max Standardization (Z-SM2S) technique can be used to improve data quality and identify null or missing values during the data preprocessing stage. Additionally, the Additive Spider Swarm Optimization Intelligence Technique (ASSOIT) is employed to determine the limitations of cyber-attack features. After that, the HDB-GANN model based on the Deep Learning (DL) algorithm can be proposed to improve the detection accuracy of financial transactions. Finally, communication records are encrypted using the Proactive-Blockchain Advanced Folding Encryption (PBAFE) model, enhancing security through master node-verified private key access. The proposed method achieves 96.27% accuracy over other methods in a wide range of simulation analyses.