Wasserstein Generative Adversarial Network with Gradient Penalty and Threshold-Enhanced for Imbalanced Panel Data for Financial Fraud Detection

Authors

  • Jordan Istiqlal Qalbi Adiba Department of Informatics, Institut Teknologi Sepuluh Nopember, Indonesia
  • Riyanarto Sarno Department of Informatics, Institut Teknologi Sepuluh Nopember, Indonesia
  • Kelly Rossa Sungkono Department of Informatics, Institut Teknologi Sepuluh Nopember, Indonesia
  • Agus Tri Haryono Department of Informatics, Institut Teknologi Sepuluh Nopember, Indonesia
  • A Min Tjoa Faculty of Computer Science, University of Vienna, Austria
  • Sang-Seok Lee Graduate School of Engineering, Tottori University, Japan

DOI:

https://doi.org/10.32890/jict2026.25.1.3

Keywords:

Fraud detection, generative adversarial network, imbalanced, oversampling, financial statement

Abstract

Financial statement fraud detection is critical to maintaining trust among investors, regulators, and analysts. However, traditional audit procedures often fail to detect anomalies effectively because they occur infrequently but can result in significant economic losses. This study proposes an oversampling approach using a modified threshold in the Wasserstein Generative Adversarial Network with Gradient Penalty (WGANGP) to enhance synthetic data variance in financial fraud detection. The financial data were collected from the financial reports of companies listed on the Indonesia Stock Exchange and were labelled according to the Balanced Scorecard framework into four categories: normal, alarm, risky, and fraud. Given the severe class imbalance, this study introduces a WGANGP model with threshold optimisation in the generator and a gradient penalty to generate high-quality synthetic samples. This study conducted general and per-entity oversampling scenarios and evaluated them using Euclidean distance, Wasserstein distance, and classification metrics. In Scenario 1, the Generative Adversarial Network (GAN) outperformed the Synthetic Minority Oversampling Technique (SMOTE) and vice versa in Scenario 2. However, in financial fraud detection, the WGANGP with enhanced thresholding improved the F1-score by 13% to 17% compared to SMOTE and five GAN-based models across thirteen classification models, including traditional, machine learning, and deep learning models. This finding suggests that optimising the threshold in WGANGP reduces variance and improves model performance. Furthermore, generating synthetic data that is very similar to actual data may not necessarily improve classification; therefore, it is necessary to test how oversampling affects subsequent stages.

References

Abueid, R., Rehman, S., & Nguyen, N. T. (2022). The impact of balanced scorecard in estimating the performance of banks in Palestine. EuroMed Journal of Business. https://doi.org/10.1108/emjb-03-2021-0047

Aftabi, S. Z., Ahmadi, A., & Farzi, S. (2023). Fraud detection in financial statements using data mining and GAN models. Expert Systems with Applications, 227. https://doi.org/10.1016/j.eswa. 2023.120144

Akinbowale, O. E., Mashigo, P., & Zerihun, M. F. (2023). Application of balance scorecard as a strategic management and performance measurement tool for cyberfraud mitigation. 7th International Conference on Business and Information Management, ICBIM 2023, 60–69. https://doi.org/10.1109/ICBIM59872.2023.10303219

Asokan, S., & Seelamantula, C. S. (2023). Euler-lagrange analysis of generative adversarial networks. Journal of Machine Learning Research, 24. http://jmlr.org/papers/v24/20-1390.html.

Baesens, B., Höppner, S., & Verdonck, T. (2021). Data engineering for fraud detection. Decision Support Systems, 150. https://doi.org/10.1016/j.dss.2021.113492

Chen, C., Shen, W., Yang, C., Fan, W., Liu, X., & Li, Y. (2023). A new safe-level enabled borderline-smote for condition recognition of imbalanced dataset. IEEE Transactions on Instrumentation and Measurement, 72. https://doi.org/10.1109/TIM.2023.3289545

Dina, A. B., Sarno, R., Anggraini, R. N. E., Haryono, A. T., & Septiyanto, A. F. (2024). Comparison of oversampling techniques in prediction judicial decisions of divorce trials in family courts. Proceeding - 2024 International Conference on Information Technology Research and Innovation, ICITRI 2024, 13–18. https://doi.org/10.1109/ICITRI62858.2024.10699016

Fadel, S., Rouaski, K., Challal, M., & Bouaicha, H. (2021). The balanced scorecard (BSC) as a multidimensional performance measurement system tool: Case the company of algeria post. Financial Markets, Institutions and Risks, 5(4). https://doi.org/10.21272/fmir.5(4).87-105.2021

Fernando, K. R. M., & Tsokos, C. P. (2022). Dynamically weighted balanced loss: Class imbalanced learning and confidence calibration of deep neural networks. IEEE Transactions on Neural Networks and Learning Systems, 33(7), 2940–2951. https://doi.org/10.1109/TNNLS.2020. 3047335

Haryono, A. T., Sarno, R., Esti Anggraini, R. N., & Sungkono, K. R. (2024). Imputation missing stock prices using generative adversarial networks and attention mechanism. 2024 2nd International Conference on Technology Innovation and Its Applications (ICTIIA), 1–6. https://doi.org/10.1109/ICTIIA61827.2024.10761233

Hashemi, S. K., Mirtaheri, S. L., & Greco, S. (2023). Fraud detection in banking data by machine learning techniques. IEEE Access, 11, 3034–3043. https://doi.org/10.1109/ACCESS.2022. 3232287

Hrishikesh, P. S., Puthussery, D., Akhil, K. A., & Jiji, C. V. (2023). Relativistic GAN using receptive field block for single image super-resolution with improved perceptual quality. 2023 11th International Symposium on Electronic Systems Devices and Computing, ESDC 2023. https://doi.org/10.1109/ESDC56251.2023.10149876

Hsin, Y. Y., Dai, T. S., Ti, Y. W., Huang, M. C., Chiang, T. H., & Liu, L. C. (2022). Feature engineering and resampling strategies for fund transfer fraud with limited transaction data and a time-inhomogeneous modi operandi. IEEE Access, 10, 86101–86116. https://doi.org/10.1109/ ACCESS.2022.3199425

Kaewninprasert, K., Chai-Arayalert, S., & Yamaqupta, N. (2024). The perspective classification of balanced scorecard with ontology technique. Journal of Information and Communication Technology, 23(3), 465–494. https://doi.org/10.32890/jict2024.23.3.4

Kim, C., Park, S., & Hwang, H. J. (2022). Local stability of Wasserstein GANS with abstract gradient penalty. IEEE Transactions on Neural Networks and Learning Systems, 33(9), 4527–4537. https://doi.org/10.1109/TNNLS.2021.3057885

Li, B., Yen, J., & Wang, S. (2024). Uncovering financial statement fraud: A machine learning approach with key financial indicators and real-world applications. IEEE Access. https://doi.org/10.1109/ ACCESS.2024.3520249

Li et al. (2023). EID-GAN: Generative adversarial nets for extremely imbalanced data augmentation. IEEE Transactions on Industrial Informatics, 19(3), 3208–3218. https://doi.org/10.1109/TII. 2022.3182781

Li, X., Da, X., Shi, W., & Liu, W. (2023, November 30). Manufacturing companies financial fraud detection based on interpretable machine learning. Proceedings of the 2nd International Conference on Public Management. https://doi.org/10.4108/eai.1-9-2023.2338768

Liao, C., & Dong, M. (2022). ACWGAN: An auxiliary classifier Wasserstein GAN-based oversampling approach for multi-class imbalanced learning. International Journal of Innovative Computing, Information and Control, 18(3), 703–721. https://doi.org/10.24507/ijicic.18.03.703

Man, C. K., Quddus, M., Theofilatos, A., Yu, R., & Imprialou, M. (2022). Wasserstein generative adversarial network to address the imbalanced data problem in real-time crash risk prediction. IEEE Transactions on Intelligent Transportation Systems, 23(12), 23002–23013. https://doi.org/10.1109/TITS.2022.3207798

Miftahushudur, T., Grieve, B., & Yin, H. (2024). Permuted KPCA and SMOTE to guide GAN-based oversampling for imbalanced HSI classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 17, 489–505. https://doi.org/10.1109/JSTARS.2023. 3326963

Mohammed, R. M. H. (2022). The impact of audit committee on financial reporting quality. Journal of Global Economics and Business, 3(11), 81–94. https://doi.org/10.31039/jgeb.v3i11.91

Naderi, M., Nabizadeh, Z., Karimi, N., Shirani, S., & Samavi, S. (2021). MSGDD-CGAN: Multi-scale gradients dual discriminator conditional generative adversarial network. ArXiv:2109.05614

Permataning Tyas, S. M., Sarno, R., Haryono, A. T., & Rossa Sungkono, K. (2023). A robustly optimised BERT using random oversampling for analysing imbalanced stock news sentiment data. ICCoSITE 2023 - International Conference on Computer Science, Information Technology and Engineering: Digital Transformation Strategy in Facing the VUCA and TUNA Era, 897–902. https://doi.org/10.1109/ICCoSITE57641.2023.10127725

Qin, R. (2021). Identification of accounting fraud based on support vector machine and logistic regression model. Complexity, 2021. https://doi.org/10.1155/2021/5597060

Saksono, L. A., & Bernardus, D. (2023). Design of balanced scorecard as a school's performance measurement. Binus Business Review, 14(2), 171–183. https://doi.org/10.21512/bbr.v14i2. 8901

Sarno, R., Dewandono, R. D., Ahmad, T., Naufal, M. F., & Sinaga, F. (2015). Hybrid association rule learning and process mining for fraud detection. IAENG International Journal of Computer Science.

Seireg, H. R., Omar, Y. M. K., El-Samie, F. E. A., El-Fishawy, A. S., & Elmahalawy, A. (2022). Ensemble machine learning techniques using computer simulation data for wild blueberry yield prediction. IEEE Access, 10, 64671–64687. https://doi.org/10.1109/ACCESS.2022.3181970

Shafqat, W., & Byun, Y. C. (2022). A hybrid GAN-based approach to solve imbalanced data problem in recommendation systems. IEEE Access, 10, 11036–11047. https://doi.org/10.1109/ ACCESS.2022.3141776

Sihwail, R., Ghamri, M. Al, & Ibrahim, D. (2024). An enhanced model of whale optimization algorithm and k-nearest neighbors for malware detection. International Journal of Intelligent Engineering and Systems, 17(3), 606–621. https://doi.org/10.22266/ijies2024.0630.

Sokolenko, L. F. (2022). Detection and evaluation of fraud during the public sector audit. Statistics of Ukraine, 97(2), 95–103. https://doi.org/10.31767/su.2(97)2022.02.10

Sunaryono et al. (2023). Optimised one-dimension convolutional neural network for seizure classification from EEG signal based on whale optimization algorithm. International Journal of Intelligent Engineering and Systems, 16(3), 310–322. https://doi.org/10.22266/ ijies2023.0630.25

Widiantoro, A. D., Mustafid, M., & Sanjaya, R. (2024). Model analytic in fintech user comment features using LDA-CNN on imbalanced data. International Journal of Intelligent Engineering and Systems, 17(4), 1079–1098. https://doi.org/10.22266/IJIES2024.0831.80

Xiuguo, W., & Shengyong, D. (2022). An analysis on financial statement fraud detection for Chinese listed companies using deep learning. IEEE Access, 10, 22516–22532. https://doi.org/10.1109/ACCESS.2022.3153478

Yin et al. (2023). Imbalanced working states recognition of sucker rod well dynamometer cards based on data generation and diversity augmentation. SPE Journal. https://doi.org/10.2118/214661-pa

Zhou, H., Sun, G., Fu, S., Wang, L., Hu, J., & Gao, Y. (2021). Internet financial fraud detection based on a distributed big data approach with node2vec. IEEE Access, 9, 43378–43386. https://doi.org/10.1109/ACCESS.2021.3062467

Downloads

Published

31-01-2026

Issue

Vol. 25 No. 1 (2026): Journal of Information and Communication Technology (JICT) Vol.25, No.1, January 2026

Section

Articles

License

Copyright (c) 2026 Journal of Information and Communication Technology

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Adiba, J. I. Q., Riyanarto Sarno, Kelly Rossa Sungkono, Agus Tri Haryono, A Min Tjoa, & Sang-Seok Lee. (2026). Wasserstein Generative Adversarial Network with Gradient Penalty and Threshold-Enhanced for Imbalanced Panel Data for Financial Fraud Detection. Journal of Information and Communication Technology, 25(1), 39-63. https://doi.org/10.32890/jict2026.25.1.3

Research impact

Harvested 2026-09-26
1 citations, from Semantic Scholar — the highest of the sources checked

Counts differ between services because each indexes a different body of literature. None of them is the whole picture.

Identifiers DOI 10.32890/jict2026.25.1.3 OpenAlex W7128513872 Semantic Scholar CorpusID 285496975 Scopus 105030849514

Most read articles by the same author(s)