Intelligent Financial Risk Prediction Framework with Optimized Hybrid Deep Learning Architecture in Supply Chain Management

A risk evaluation on the supply chain model is developed using a deep learning network to investigate the risk in supply chain management, encourage cooperation among businesses, and assure the good and normal supply chain's development. Many models have demonstrated that the Support Vector Machine (SVM) outperforms alternative techniques for classifying and making predictions. Therefore, this paper considers using a combination of SVM with a deep learning scheme to establish an evaluation model. This paper provides efficient performance on evaluating the financial risks by performing (a) Risk data collection and (b) Financial risk evaluation. The essential financial data from supply chain management is collected for making the decision on the financial risk. The collected financial data are utilized in the risk evaluation model, where Adaptive Serial Cascaded Autoencoder with Support Vector Machine (ASCA- SVM) is developed to assess the risks in the SCM. Here, the parameter in the developed ASCA-SVM is tuned to enhance the evaluation efficiency with the support of the Dragonfly Algorithm (DA). Experimental analysis reveals that the proposed approach attains better performance with the benchmark dataset when compared with the conventional risk evaluation techniques.