Predictive Analytics

  1. Ensemble Methods
    1. Ensemble Learning Principles
      1. Diversity in Ensemble Members
        1. Bias-Variance Decomposition
          1. Ensemble Size Considerations
            1. Combination Strategies
            2. Bagging Methods
              1. Bootstrap Aggregating
                1. Bootstrap Sampling
                  1. Aggregation Strategies
                    1. Variance Reduction
                    2. Random Forests
                      1. Random Feature Selection
                        1. Out-of-Bag Error Estimation
                          1. Feature Importance Calculation
                            1. Hyperparameter Tuning
                            2. Extra Trees
                              1. Extremely Randomized Trees
                                1. Random Threshold Selection
                                  1. Computational Efficiency
                                2. Boosting Methods
                                  1. AdaBoost
                                    1. Adaptive Weight Adjustment
                                      1. Weak Learner Requirements
                                        1. Error Rate Minimization
                                        2. Gradient Boosting
                                          1. Gradient Descent Framework
                                            1. Loss Function Optimization
                                              1. Residual Fitting
                                              2. Advanced Boosting Algorithms
                                                1. XGBoost
                                                  1. Regularization Features
                                                    1. Parallel Processing
                                                      1. Missing Value Handling
                                                      2. LightGBM
                                                        1. Leaf-wise Tree Growth
                                                          1. Categorical Feature Support
                                                            1. Memory Efficiency
                                                            2. CatBoost
                                                              1. Categorical Feature Processing
                                                                1. Ordered Boosting
                                                                  1. Overfitting Reduction
                                                              2. Stacking and Blending
                                                                1. Meta-learning Approach
                                                                  1. Base Model Selection
                                                                    1. Meta-model Training
                                                                      1. Cross-validation Stacking
                                                                        1. Blending Strategies
                                                                        2. Voting Methods
                                                                          1. Hard Voting
                                                                            1. Soft Voting
                                                                              1. Weighted Voting
                                                                                1. Dynamic Voting

                                                                              Previous

                                                                              4. Classification Modeling

                                                                              Go to top

                                                                              Next

                                                                              6. Neural Networks and Deep Learning