ATM Fraud Detection Using Outlier Detection

A fraud detection model that applies accounts' behavior features and outlier detection methods are proposed. Given a set of transactions, the accounts are grouped into two groups according to the location that the transactions took place, i.e., 'local-only' and `has-abroad'. A feature is extracted to reflect the normal behavior of the accounts in each group. Only known legitimate transactions are used to extract a set of features for representing a legitimate behavior. An unknown transaction is classified either normal or fraud using an outlier detection. The experimental result shows that the proposed feature with an Isolation Forest outlier detection technique is able to detect all fraud transactions.