Deep learning model based research on anomaly detection and nancial fraud identi cation in corporate nancial reporting statements

Financial frauds, often executed through asset transfers and pro t in ation, aim to reduce taxes and secure credits. To enhance the accuracy and e ciency of accounting data auditing, this study proposes an anomaly detection scheme based on a deep autoencoder neural network. Financial statement entries are extracted from the accounting information system, and global and local anomaly features are de ned based on the attribute values of normal and fraudulent accounts, corresponding to individual and combined anomaly attribute values. The AE network is trained to identify anomalies using account attribute scores. Results demonstrate classi cation accuracies of 91.7%, 90.3%, and 90.9% for sample ratios of 8:2, 7:3, and 6:4, respectively. The precision, recall, and F1 score reach 90.85%, 90.77%, and 90.81%, respectively. Training takes 95.81ms, with recognition classi cation requiring only 0.02ms. The proposed deep neural network achieves high recognition accuracy and speed, signi cantly improving the detection of nancial statement anomalies and fraud. © 2024 The Author(s).