Fit IsolationForest on normal-dominated data; plot the score distribution; review the top alerts.
Concepts: Random partitioning; isolation depth; anomaly score; contamination
Project A: Score all transactions; top 500 alerts → how many are real fraud (hidden labels)?
Project B: Daily features (mean, peak, min, peak hour, temperature, weekday) → unusual days.
Notes, key code, what to skip and 2 curated resources are inside the platform.Join the waitlist and we'll build you a roadmap that includes this topic only if your goal needs it.