Explainable Team Formation by Integrating Skill Evolution and High-Order Collaboration

Explainable team formation has gained significant attention for its ability to enhance decision-makers’ trust and team effectiveness. The dynamic evolution of experts’ skills and their high-order collaborative relationships critically impact team performance and recommendation credibility. However, existing methods fail to capture the temporal decay of skill proficiency or model complex collaborative structures beyond direct relationships, while lacking sufficient justification for their recommendations. To address these limitations, we propose a novel Explainable Team Formation Model (ETFM) that incorporates skill evolution and higher-order collaboration. Specifically, we first construct a skill evolution-aware collaboration network enhanced with temporal decay factors to dynamically track proficiency levels. We then develop a hypergraph neural network-based representation learning method to capture latent synergy patterns from high-order relationships. Finally, we define two quantitative metrics (“skill familiarity” and “collaboration tightness”) to generate transparent justifications. Extensive experiments on four real-world datasets demonstrate that our approach significantly outperforms state-of-the-art baselines across multiple evaluation metrics.