Anomaly Detection in General Ledger Data: Results from a Hybrid Approach

The paper explores a hybrid approach that combines traditional Journal Entry Tests (JETs) with machine learning techniques to enhance anomaly detection in general ledger data. It presents specialized models designed to improve the accuracy and validity of detected anomalies, thereby aiming to reduce false positives and increase audit efficiency. Experiments are conducted using synthetic data that includes both normal and anomalous journal entries.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.