Find Unusual Segments in Your Data with Subgroup Discovery

Patient rule induction method finds 35% better segments than previously reported

Inspired by an in-depth Medium article [1] with a case study on identifying bank customer segments with high churn reduction potential, this story explores a similar challenge through the lens of subgroup discovery methods [2]. Intrigued by the parallels, I applied a subgroup discovery approach to the same dataset and uncovered a segment with a 35% higher churn reduction potential — a significant …

article author bank case case study challenge churn customer data discovery insights-and-data interpretability machine learning medium patient recraft story study through