Find out the latest evidence on the use of undersampling, oversampling and SMOTE for imbalanced data and whether the Python library imblearn is still relevant.
ADASYN: Adaptive Synthetic Sampling for Imbalanced Datasets
Find out why you should NOT use ADASYN in 2025 to handle data imbalance, what the hype behind ADASYN was, and what you could do instead to boost model performance.
The Complete Guide to Platt Scaling
Learn about calibration in machine learning using Platt scaling. Find out how it works and how to apply it in Python using Scikit-learn.
SMOTE in Python and whether you should still use it in 2025
Learn how to implement SMOTE in Python and whether you should still be using it to work with imbalanced datasets in 2025.
ROC-AUC Analysis – A Deep Dive
Ultimate guide for mastering ROC-AUC analysis—learn to create, interpret, and apply it in Python with practical examples.
Probability Calibration in Machine Learning: Enhancing Model Usability
Learn probability calibration in machine learning: importance, methods, and best practices for more reliable probability estimates.
A Data Scientist’s Guide to Balanced Accuracy
Discover the balanced accuracy’s advantages over traditional accuracy and learn how to implement it in Python.
Precision Recall Curves
The ultimate guide to Precision-Recall curves—what they are, when to use them, and how to plot and interpret them with Python.
Class Imbalance in Machine Learning
Contrary to what you’ll read online or get from ChatGPT, class imbalance is NOT the problem. How you handle it is.
Cost-Sensitive Learning: Beyond the Accuracy in Imbalanced Classification
Find out what cost-sensitive learning is and how to implement it with Python.