Unverified60% confidenceFactExact time
LIME explains individual predictions by generating a perturbed dataset around the input sample to train a local linear approximation model, while SHAP provides feature attribution based on game-theoretic Shapley values; both are model-agnostic techniques.
2
Sources
60%
Confidence
Long-term
Relevance
9/4/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLIME通过在目标样本周围构造扰动数据集，训练局部可解释的线性模型来近似复杂模型的局部行为79% similarUnverifiedSHAP值通过博弈论中的沙普利值方法量化每个特征对单一预测结果的贡献度，LIME通过在预测点附近构建局部线性近似模型解释个案决策78% similarVerifiedSHAP (SHapley Additive exPlanations) is a mainstream machine learning model explanation framework whose theoretical basis derives from Shapley values in game theory.77% similarUnverifiedSHAP的核心思想是将每个特征对模型预测的贡献量化，使得所有特征贡献之和等于模型预测值与基线值的差74% similarVerifiedSHAP基于博弈论中的Shapley值，能够量化每个特征对最终预测结果的贡献度，是目前工业界应用最广泛的解释性工具之一74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/855814API
curl https://kongchang.com/api/v1/knowledge/claims/855814MCP
get_claim(id=855814)