Document Type : Research Article
Authors
1
Department of Industrial Engineering, Faculty of Engineering, College of Farabi, University of Tehran, Tehran, Iran
2
Information Science and Epistemology, Faculty of Humanities, University of Qom, Qom, lran
10.22034/ahr.2026.2089902.1344
Abstract
As corruption patterns become more complex within the digital economy ecosystem, traditional oversight methods are no longer sufficiently effective. Utilizing a scientometric approach and analyzing 3,212 scientific records, this study aims to map a comprehensive knowledge landscape at the intersection of AI and anti-corruption to identify scientific hubs, emerging trends, and optimal regulatory models. Data were extracted from the Web of Science citation database. Data analysis was conducted using VOSviewer, UCINET, and Excel to visualize the co-occurrence network of keywords and thematic clustering. Publication trend analysis indicates that 2024 represents the peak of growth in this field with 602 scientific records. China is the primary hub of scientific production with 722 records, while Iran ranks 16th globally with 68 records. Furthermore, the findings reveal that journal articles dominate with 2,023 records, and English is the predominant language with a 98.93% share. At the institutional level, the State University System of Florida with 90 records and the Chinese Academy of Sciences with 65 records were identified as the most active scientific centers worldwide in this research intersection. The journal IEEE Access serves as the leading driver among specialized publications, having published 152 documents. Thematic findings confirm that Computer Science forms the core of this research with 2,264 records. Analysis of the knowledge maps led to the identification of five strategic clusters covering areas such as public procurement oversight, financial statement fraud detection, tracking financial flows in blockchain and cryptocurrencies, and explainable AI. These results demonstrate that the global focus has shifted from traditional auditing toward high-precision algorithmic oversight and interpretable models for legal accountability. The analyses indicate that the anti-corruption paradigm has transitioned from traditional methods toward data-driven governance and algorithmic monitoring. By focusing on machine learning, corruption bottlenecks in public procurement and financial markets can be identified with high accuracy.
Keywords
Subjects