A bank notices unusual activity in a customer’s account, such as multiple large withdrawals in a short period. What is the most appropriate technique to detect fraud in such cases?

Anomaly detection identifies data points that deviate significantly from the norm. In fraud detection, it flags unusual transactions, such as large withdrawals or geographically improbable activity, for further investigation. Techniques like isolation forests or Z-scores are commonly used. Option A : Logistic regression predicts binary outcomes but isn’t specialized for anomalies. Option C : Clustering groups similar data but doesn’t detect deviations effectively. Option D : Decision trees classify data but are not optimal for real-time anomaly detection. Option E : A/B testing is unrelated to fraud detection.