Talent Prediction Based on Switching Strategy Algorithm of Residual Grey Model

The surge in demand for medical equipment necessitates accurate prediction of talent demand in the medical industry, becoming a crucial support for the strategic planning of talent and human resources in universities. However, the difficulty in effectively collecting talent data in the medical equipment industry results in limited availability of talent demand data. At the same time, existing research focuses only on the linear or nonlinear single relationship of talent demand sequence data, thereby reducing the accuracy and robustness of predictions. In this paper, we propose a Switching Strategy Algorithm based on Residual Grey Model (SSA-RGM) to address the above issue. SSA-RGM is a hybrid model that includes a novel residual grey model, commonly used simple linear regression model, and a switching strategy based on optimal fitness. In this model, we introduce a residual operation based on the grey model and, for the first time, incorporate a penalty factor to mitigate the impact of linear shifts in small-sample data within nonlinear systems. Furthermore, we adopt the principle of optimal fitness to select the best model in the hybrid model. Experimental results on six real-world datasets show that SSA-RGM achieves the best talent prediction performance in various settings.