A novel multi-attribute group decision-making method for talent evaluation using heterogeneous data weighting and an extended cloud-VIKOR model

In the context of globalization and the knowledge economy, universities’ competitiveness heavily depends on their talent pool. Evaluating university talent is a complex but essential task. A fair and well-designed evaluation framework can enhance teachers’ satisfaction and motivation, ultimately improving educational quality. This paper introduces an innovative Multi-Attribute Group Decision-Making method for talent ranking. First, an indicator system involving both quantitative and qualitative indicators is established. The evaluation of qualitative indicators is performed by integrating heterogeneous data, group decision-making techniques and Normal Cloud Models (NCMs). Heterogeneous data characterized by various uncertainties are transformed into NCMs, and the Wasserstein distance is employed to quantify the differences between two NCMs. Subsequently, a novel Group Heterogeneous Data Direct Weight Method is proposed, utilizing Uncertain Degree and Difference Degree to determine the weights of experts during both qualitative indicator evaluation and indicator weighting. Finally, an NCM extended VlseKriterijumska Optimizacija I Kompromisno Resenje (NCM-VIKOR) method is proposed to rank alternative talents. A case study validates the effectiveness and practicality of the proposed approach. The result analysis, sensitivity analysis, comparative analysis, and superiority analysis demonstrate the rationality, robustness, uniqueness, and strengths. The proposed method can process a wider range of data types, generate more stable and informative results. Furthermore, it handles and propagates uncertainty stemming not only from indicator evaluations but also from indicator weights.