Risk prediction of enterprise human resource management based on deep learning

Abstract

BACKGROUND: The efficiency and accuracy of risk prediction in traditional enterprise human resource management (HRM) cannot meet practical needs. In response to this deficiency, this study proposes an enterprise HRM risk intelligent prediction model based on deep learning. METHODS: Two tasks were completed in this study. First, based on the existing research results and the current status of enterprise HRM, the HRM risk assessment system is constructed and streamlined. Second, for the defects of Back Propagation Neural Network (BPNN) model, Seagull Optimization Algorithm (SOA) is used to optimize it. The Whale Optimization Algorithm (WOA) is introduced to promote the SOA for its weak global search capability and its tendency to converge prematurely. RESULTS: By simplifying the HR risk assessment system and optimizing the BPNN using the SOA algorithm, an intelligent HRM risk prediction model based on the ISOA-BPNN was constructed. The results show that the error value of the ISOA-BPNN model is 0.02, the loss value is 0.50, the F1 value is 95.7%, the recall value is 94.9%, the MSE value is 0.31, the MAE value is 8.4, and the accuracy is 99.53%, both of which are superior to the other two models. CONCLUSIONS: In summary, the study of the HRM risk intelligent prediction model constructed based on ISOA-BPNN has high accuracy and efficiency, which can effectively achieve HRM risk intelligent prediction and has positive significance for enterprise development.