This paper explores feature weighting for clustering in unsupervised learning.New weighting strategies based on SHAP are presented.Empirical evaluations show that SHAP-based feature weighting improves clustering quality.The results offer insights for practical applications.