Data Mining and Knowledge Discovery

  1. Model Interpretation and Explainability
    1. Interpretability Concepts
      1. Global vs Local Interpretability
        1. Model-Agnostic vs Model-Specific Methods
          1. Intrinsic vs Post-Hoc Interpretability
          2. Feature Importance Analysis
            1. Permutation Importance
              1. Feature Contribution Methods
                1. Partial Dependence Plots
                  1. SHAP Values
                  2. Local Explanation Methods
                    1. LIME Algorithm
                      1. Local Surrogate Models
                        1. Counterfactual Explanations
                        2. Global Explanation Methods
                          1. Model Distillation
                            1. Rule Extraction
                              1. Global Surrogate Models
                              2. Visualization Techniques
                                1. Decision Tree Visualization
                                  1. Feature Space Visualization
                                    1. Cluster Visualization
                                      1. Association Rule Visualization
                                        1. Network Visualization
                                        2. Knowledge Representation
                                          1. Rule-Based Representations
                                            1. Semantic Networks
                                              1. Ontologies
                                                1. Report Generation

                                              Previous

                                              9. Model Evaluation and Validation

                                              Go to top

                                              Next

                                              11. Deployment and Production Systems