MACHINE LEARNING AS THE AUDITOR OF THE FUTURE: AUTOMATED PREDICTION OF ACCOUNTING RISKS AND IRREGULARITIES
DOI:
https://doi.org/10.56238/revgeov16n5-250Keywords:
Accounting 4.0, Predictive Auditing, Machine Learning, Accounting Risks, Accounting IrregularitiesAbstract
The digital transformation and Accounting 4.0 have driven the adoption of advanced data analytics techniques in auditing and accounting control, particularly through the use of machine learning algorithms. This study aims to assess the predictive capability of machine learning models in identifying accounting risks and irregularities in Brazilian publicly traded companies, contributing to the consolidation of predictive auditing as an evolution of traditional audit practices. To this end, the analysis is based on economic and financial data, performance indicators, earnings quality measures, and textual features extracted from notes to the financial statements, covering the period from 2010 to 2024. Several supervised algorithms, including ensemble learning models, were trained and evaluated using cross-validation and performance metrics such as accuracy, F1-score, and area under the ROC curve. The results show that ensemble-based models outperform traditional statistical methods in predicting accounting irregularities, especially when numerical information is combined with textual disclosures, reinforcing the potential of artificial intelligence as a support tool for continuous auditing and risk management. From a theoretical perspective, the study advances the understanding of how Accounting 4.0, predictive auditing, and machine learning can be integrated into a coherent analytical framework. From a practical standpoint, it provides insights for the development of automated accounting monitoring systems that are more efficient, preventive, and aligned with contemporary demands for governance and transparency.
Downloads
References
ALLES, M.; BRENNAN, G.; KOGAN, A. Continuous auditing: The state of the art and future directions. Journal of Information Systems, Sarasota, v. 32, n. 3, p. 1–20, 2018. doi:10.2308/isys-52089.
APPELBAUM, D. et al. Audit analytics and continuous audit: Opportunities and challenges. Accounting Horizons, Sarasota, v. 31, n. 3, p. 63–81, 2017.
doi:10.2308/acch-51765.
BAO, Y.; LI, H. Detecting accounting fraud using machine learning approaches: A systematic literature review. Journal of Accounting Literature, Greenwich, v. 48, p. 1–22, 2022. doi:10.1016/j.acclit.2021.100459.
BROWN-LIBURD, H.; VASARHELYI, M. A. Big Data and audit evidence. Journal of Emerging Technologies in Accounting, Sarasota, v. 12, n. 1, p. 1–16, 2015. doi:10.2308/jeta-10468.
CARVALHO, E.; REZENDE, A. J. Data analytics e os novos caminhos da auditoria interna: percepções e desafios. Revista Universo Contábil, Blumenau, v. 17, n. 3, p. 30–48, 2021. doi:10.4270/ruc.2021321.
COSTA, A. M.; LIMA, R. S.; ALCÂNTARA, J. L. Detecção automatizada de anomalias contábeis por meio de algoritmos de machine learning: evidências em bases financeiras brasileiras. Revista Brasileira de Contabilidade, Brasília, v. 51, n. 242, p. 45–62, 2022.
CRESWELL, J. W.; CRESWELL, J. D. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 5. ed. Thousand Oaks: Sage, 2018.
DAMASCENA, L.; ALMEIDA, J. E. F.; TAVARES, E. Big Data e profissional contábil: desafios e oportunidades na era da informação. Revista Contemporânea de Contabilidade, Florianópolis, v. 19, n. 48, p. 75–92, 2022. doi:10.5007/2175-8069.2022.e83825.
DECHOW, P. M.; GE, W.; SCHRAND, C. M. Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of Accounting and Economics, Amsterdam, v. 50, n. 2–3, p. 344–401, 2010. doi:10.1016/j.jacceco.2010.09.001.
HAIR, J. F. et al. Multivariate Data Analysis. 8. ed. New York: Cengage, 2019.
HE, W.; CASEY, R. J. Earnings management and financial distress: An international perspective. Journal of Accounting, Auditing & Finance, London, v. 36, n. 3, p. 572–599, 2021. doi:10.1177/0148558X19877096.
KUHN, M.; JOHNSON, K. Feature Engineering and Selection: A Practical Approach for Predictive Models. Boca Raton: CRC Press, 2021.
KNECHEL, W. R.; SALTERIO, S.; BALLA, B. Auditing: Theory and Practice. New York: Routledge, 2016.
LIN, C.; HWANG, Y. Machine learning-based detection of financial anomalies: A comprehensive review. International Journal of Accounting Information Systems, Amsterdam, v. 51, p. 100619, 2023. doi:10.1016/j.accinf.2023.100619.
MARTINS, A. L. S.; MONTEIRO, R. P. Contabilidade 4.0 e o futuro da profissão contábil: competências, desafios e perspectivas tecnológicas. Revista Contemporânea de Contabilidade, Florianópolis, v. 20, n. 54, p. 89–108, 2023. doi:10.5007/2175-8069.2023.e104589.
PUCKETT, A.; BRODY, R. Artificial intelligence and its impact on corporate governance: A governance analytics perspective. Journal of Forensic & Investigative Accounting, New Orleans, v. 13, n. 3, p. 415–437, 2021.
RASHID, A.; WANG, S. Predictive analytics in accounting and auditing: A review and research agenda. International Journal of Accounting Information Systems, Amsterdam, v. 50, p. 100610, 2023. doi:10.1016/j.accinf.2023.100610.
ROSMAN, P.; SOTO, M.; KHASHMAN, A. Digital transformation and its impact on accounting and auditing education. Innovations in Education and Teaching International, London, v. 58, n. 5, p. 620–633, 2021. doi:10.1080/14703297.2020.1815714.
SCHMIDT, P. et al. The digital transformation of the accounting profession: Implications for practitioners and educators. Accounting Education, London, v. 29, n. 5, p. 439–466, 2020. doi:10.1080/09639284.2020.1754800.
SHMUELI, G.; KOPPIUS, O. Predictive analytics in information systems research. MIS Quarterly, Minneapolis, v. 35, n. 3, p. 553–572, 2011.
SILVA, T. B.; ARAÚJO, L. F.; MENDONÇA, M. F. Data analytics e auditoria interna: oportunidades e barreiras em organizações brasileiras. Revista de Educação e Pesquisa em Contabilidade, Brasília, v. 16, n. 4, p. 489–507, 2022. doi:10.17524/repec.v16i4.3030.
TUNG, B.; VASARHELYI, M. A. The evolution from traditional auditing to predictive auditing: Implications for the profession. International Journal of Disclosure and Governance, London, v. 20, p. 45–60, 2023. doi:10.1057/s41310-022-00166-7.
VASARHELYI, M. A.; HALPER, F. The continuous audit of online systems. Auditing: A Journal of Practice & Theory, Sarasota, v. 10, n. 1, p. 110–125, 1991.
VASARHELYI, M. A.; ROEDEL, K. The future of continuous auditing: Artificial intelligence and real-time assurance. Journal of Emerging Technologies in Accounting, Sarasota, v. 19, n. 1, p. 87–103, 2022. doi:10.2308/JETA-2021-054.
WARREN, J. D.; MOFFITT, K. C.; BYRNES, P. How Big Data will change accounting. Accounting Horizons, Sarasota, v. 29, n. 2, p. 397–407, 2015. doi:10.2308/acch-51069.
