Economic Benefit Risk Prediction Based on Feature Selection and Deep Learning Model
The combination of big data,cloud computing and artificial intelligence technologies has significantly enhanced the capability of enterprise financial data processing.In order to improve the accuracy and reliability of financial risk prediction for small and medium-sized enterprises(SMEs),an financial risk prediction framework based on multi-verse optimization(MVO)algorithm and bidirectional gated recurrent units(BiGRU).Initially,complex financial data are subjected to feature normalization,followed by the selection of the optimal feature subset using the MVO algorithm.Subsequently,the evaluation of economic benefit risk for SMEs is accomplished using the BiGRU deep learning model.The sequential model-based algorithm configuration(SMAC)is employed to perform parameter tuning for the BiGRU model,optimizing its parameter configuration to enhance model performance and generalization ability.The SMAC algorithm automatically searches for the best combination of parameters in the parameter space,thereby identifying the optimal model configuration.Experimental results demonstrate that the proposed hybrid model exhibits higher accuracy and predictive capability in the task of predicting financial risk for SMEs,outperforming similar state-of-the-art methods,thereby confirming the potential and importance of feature selection and deep learning models in economic benefit risk analysis.
financial risk predictiondeep learningfeature selectionmulti-verse optimizationbidirectional gated recurrent units
万方数据
