Explainable Artificial Intelligence

3.

Intrinsically Interpretable Models

3.1.

3.1.1.

Linear Regression

3.1.1.1.

Coefficient Interpretation

3.1.1.2.

Statistical Significance

3.1.1.3.

Assumptions and Limitations

3.1.1.4.

Regularization Effects

3.1.2.

Logistic Regression

3.1.2.1.

Odds Ratios and Log-Odds

3.1.2.2.

Probability Interpretation

3.1.2.3.

Feature Impact Analysis

3.1.3.

Generalized Linear Models

3.1.3.1.

3.1.3.2.

Exponential Family Distributions

3.1.4.

Interpreting Model Parameters

3.1.4.1.

Feature Scaling Effects

3.1.4.2.

Multicollinearity Considerations

3.1.4.3.

Confidence Intervals

3.2.

Tree-Based Models

3.2.1.

3.2.1.1.

Tree Structure and Components

3.2.1.2.

Split Criteria and Information Gain

3.2.1.3.

Pruning and Complexity Control

3.2.2.

Visualizing Decision Paths

3.2.2.1.

Tree Diagrams and Representations

3.2.2.2.

Path Tracing for Individual Predictions

3.2.2.3.

Decision Rules Extraction

3.2.3.

Feature Importance in Trees

3.2.3.1.

Split-Based Importance

3.2.3.2.

Permutation Importance

3.2.3.3.

Depth and Frequency Considerations

3.3.

Rule-Based Systems

3.3.1.

Decision Rules and Rule Lists

3.3.1.1.

IF-THEN Rule Structure

3.3.1.2.

Rule Ordering and Priority

3.3.1.3.

Rule Coverage and Overlap

3.3.2.

Association Rule Mining

3.3.2.1.

Support, Confidence, and Lift Metrics

3.3.2.2.

Apriori Algorithm

3.3.2.3.

FP-Growth Algorithm

3.3.3.

Rule Extraction from Data

3.3.3.1.

Supervised Rule Learning

3.3.3.2.

Unsupervised Pattern Discovery

3.4.

Generalized Additive Models

3.4.1.

Additive Structure

3.4.1.1.

Component Functions

3.4.1.2.

Smoothing Techniques

3.4.2.

Shape Function Interpretation

3.4.2.1.

Partial Effect Visualization

3.4.2.2.

Nonlinear Feature Relationships

3.4.3.

Interaction Terms

3.4.3.1.

Two-Way Interactions

3.4.3.2.

Higher-Order Interactions

3.5.

Probabilistic Models

3.5.1.

3.5.1.1.

Feature Independence Assumption

3.5.1.2.

Probabilistic Feature Contributions

3.5.1.3.

Class Conditional Probabilities

3.5.2.

Bayesian Networks

3.5.2.1.

Graphical Model Structure

3.5.2.2.

Conditional Independence

3.5.2.3.

Inference and Explanation

3.6.

Distance-Based Models

3.6.1.

k-Nearest Neighbors

3.6.1.1.

Instance-Based Reasoning

3.6.1.2.

Neighbor Influence Analysis

3.6.1.3.

Distance Metric Effects

3.6.2.

Prototype-Based Models

3.6.2.1.

Representative Examples

3.6.2.2.

Prototype Selection Methods

Previous

2. Taxonomy and Classification of XAI Methods

Go to top

Next

4. Post-hoc Explanation Methods