Machine learning
SHAP (SHapley Additive exPlanations)
SHAP 是一种模型解释方法，由 Scott Lundberg 和 Su-In Lee 于 2017 年提出，它利用合作博弈论中的 Shapley 值来衡量每个特征对个体预测的贡献程度，从而使黑箱机器学习模型的输出具有可解释性。它同时支持全局解释（整体特征重要性）和局部解释（特定预测值为何如此）。
阅读完整方法
仅限会员
登录使用免费账户登录即可阅读本节。
Method map
The neighbourhood of related methods — select a node to explore.
来源
- Lundberg, S.M. & Lee, S.-I. (2017). A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems, 30, 4766–4777. link ↗
如何引用本页
ScholarGate. (2026, June 1). SHAP (SHapley Additive exPlanations). ScholarGate. https://scholargate.app/zh/machine-learning/shap-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
Compare side by side →