Enhancing corporate governance via machine learning and statistical tools for fraud detection

Abstract

Background: Recent trends in machine learning and statistical techniques have revolutionised traditional auditing practices and unveiled new horizons to enhance corporate governance practices.

Objectives: This article proposes a hybrid model by combining statistical techniques, such as Benford’s Law and the Beneish M-Score, with machine learning algorithms to detect fraud. Integration of all the methodologies results in a broad, flexible framework for the identification of irregularities and possible fraudulent activities within financial datasets.

Method: The research addresses how these advanced tools meet the gaps in traditional auditing practices, thus enabling a more refined approach towards fraud detection.

Results: Empirical findings show that this integrated model will improve detection rates, thus strengthening governance structures and promoting transparency within organisations.

Conclusion: Major findings suggest that while machine learning algorithms are effective in improving the identification of complex fraud patterns, statistical methods prove to be effective in preliminary screening.

Contribution: The article ends with a discussion on implications for auditors and corporate governance structures along with future research recommendations and applications by the industry.