Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JIMEB: Jurnal Ilmiah Manajemen, Ekonomi dan Bisnis
Santoso, Amelia Ratna
Unknown Affiliation
Economics, Econometrics & Finance

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Machine Learning-Based Forensic Analytics for the Detection of Financial Statement Fraud Santoso, Amelia Ratna; Prasetyo, Budi
Jurnal Ilmiah Manajemen, Ekonomi dan Bisnis Vol. 4 No. 1 (2025): JIMEB
Publisher : Universitas Sains dan Teknologi Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/h1hep018

Abstract

The digital transformation in accounting has not only made financial reporting faster but also opened avenues for data manipulation and fraudulent reporting. As a result, there is a need to apply machine learning-based forensic analytics to scale up both objective and contextual anomaly detection. The primary purpose of the research is to introduce a fraud detection model for the financial statements of publicly listed companies in Indonesia by comparing the effectiveness of three major algorithms: Logistic Regression, Random Forest, and XGBoost. The 240 observations from 40 companies for the 2018–2023 period were analyzed using an explanatory quantitative approach, with SMOTE balancing and Beneish M-Score validation methods. The findings indicate that among all models tested, XGBoost stands out as the best, achieving the highest accuracy of 92.1% and an AUC of 0.947, whereas leverage, profit growth, and auditor turnover are identified as the main predictors of fraud risk. The application of this combination of forensic accounting and machine learning has led to the conclusion that it is possible to detect fraud with much higher accuracy than with conventional auditing techniques. One of the most important aspects of this research is the use of interpretable predictive analytics in developing countries. This means the model could serve as a data-based early warning system for auditors and regulators, thereby improving transparency and corporate governance in Indonesia.
Co-Authors Budi Prasetyo