Computer engineering frameworks: A bibliometric analysis of IT risk and audit systems

This study presents a bibliometric analysis of computer engineering frameworks applied to IT risk management and intelligent audit systems, based on 393 publications from the Web of Science Core Collection (2024–2025). Findings reveal a rapidly evolving, highly collaborative research domain with an average of 4.66 coauthors per document and 15 international co-authorship. Thematic mapping identifies four core clusters: (1) representation learning and feature engineering, (2) deep learning and detection systems, (3) computational modeling and intelligent surveillance, and (4) real-time tracking and continual learning. The integration of graph-based learning, deep clustering, and transformer architectures enhances audit transparency, anomaly detection, and system resilience. Despite methodological advances, gaps persist in aligning intelligent systems with governance mechanisms and risk controls. The study offers a structured overview of emerging trends, thematic evolution, and future directions, providing valuable insights for researchers, auditors, and system designers aiming to develop secure, adaptive, and intelligent audit infrastructures grounded in robust computer engineering principles.