What is your approach to identifying anomalies or outliers in a large dataset of consumer financial transactions?
