Imbalanced Classification

Go to main | Course Page

Section 1: Introduction

  1. Problem with Imbalanced Classes
  2. Resources
  3. Build and examine baseline
  4. Download Resources

Section 2: Sampling Based Approaches

  1. Approach 1: Upsampling the minority class
  2. Approach 2: Downsampling the majority class
  3. Approach 3: Cluster Centroids - Theory
  4. Code Demo
  5. Approach 4: SMOTE Algorithm Theory
  6. Code Demo
  7. Approach 5: ADASYN Algorithm Theory
  8. Code Demo
  9. Approach 6: Near Miss
  10. Approach 7: Tomek Links
  11. Code Demo

Section 3: Class Sensitive Learning Approaches

  1. Class Sensitive Learning Theory
  2. Code Demo
  3. Class Weighted XGBoost and Demo

Section 4: Advanced Methods

  1. One Class Classification Theory
  2. One Class using SVM
  3. Mahalanobis Distance for One Class Classification
  4. Isolation Forest Demo
  5. Data Leakage - A sneaky surprisingly hard to detect example
  6. Edited Nearest Neighbour based example
  7. Probability Threshold Tuning
Report abuse