Design and Development of a Fraud Detection and Identification Model with a Special Emphasis on Internal Auditing and Artificial Intelligence in the Context of Digital Transformation

Design and Development of a Fraud Detection and Identification Model with a Special Emphasis on Internal Auditing and Artificial Intelligence in the Context of Digital Transformation

Deniyal abdolvand1

1) M.A. in Auditing

IOI Code : ABDE-ABFDHB
Publication : 6th International Conference on Advanced Research in Management And Humanities(huconf.com)
Abstract :
    This study was conducted within the framework of digital transformation. It is an applied research project employing a descriptive–analytical approach, and its field component is based on semi structured interviews with 23 experts, including financial managers, internal auditors, and university faculty members. The qualitative data were analyzed using thematic analysis following Braun and Clarke’s six step framework. In the initial stage, 90 primary codes were extracted from interview transcripts and research documents. These codes were then categorized into 22 basic themes and, after systematic refinement and review, consolidated into 6 organizing themes. Ultimately, the organizing themes were integrated into 6 overarching themes representing the core dimensions of the proposed model. These dimensions include developing an internal audit model focused on fraud detection, leveraging data analytics and technology in auditing, improving disclosure quality, strengthening anti fraud culture and organizational accountability, utilizing artificial intelligence in financial data analysis, and enhancing information security through encryption and systemic controls. The findings show that integrating internal audit capabilities with advanced technologies such as data mining, machine learning algorithms, behavioral financial analytics, and intelligent systems can significantly improve the accuracy and speed of fraud detection and provide a more comprehensive framework for fraud risk management. Furthermore, the reliability of the qualitative analysis was evaluated using Holsti’s coefficient, Scott’s Pi, Cohen’s Kappa, and Krippendorff’s Alpha, all indicating satisfactory reliability and validity of the coding process. Based on the results, the final model—comprising 90 initial codes, 22 basic themes, and 6 overarching themes—can serve as a unified framework for enhancing internal audit performance and establishing an intelligent fraud detection system within organizations.

Keywords : internal auditing fraud artificial intelligence digital transformation fraud detection