EUVIMEDEUVIMED entdecken
Crossref

AI-Driven Anomaly Detection for Fraud Prevention in Project Monitoring

Mehmet Tahir Sandıkkaya, Onur Behzat Tokdemir

Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi · 30.12.2025 · 16 (4) · 1103-1111

DOI: 10.24012/dumf.1656802 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, online erschienen am 30.12.2025. Verfasst von Mehmet Tahir Sandıkkaya und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 36 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate tracking of project progress is crucial for timely delivery, cost control, and fraud prevention. Issues in progress reporting, whether due to real mistakes, employee inefficiencies, or internal threats, present considerable risks to major projects. This study aims to examine statistical and machine learning techniques to identify data inconsistencies, fraudulent reporting, and other anomalies in project tracking. Utilizing a dataset of 118 weekly snapshots, including genuine and tainted data, this research assesses the effectiveness of the interquartile range, isolation forest, and an ensemble approach in detecting anomalies. The results underscore the strengths and weaknesses of statistical and machine learning models while proposing an optimal detection framework for effective project management.

Zitieren

APA

Sandıkkaya, M. T., & Tokdemir, O. B. (2025). AI-Driven Anomaly Detection for Fraud Prevention in Project Monitoring. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi, 16(4), 1103-1111. https://doi.org/10.24012/dumf.1656802

Vancouver

Sandıkkaya MT, Tokdemir OB. AI-Driven Anomaly Detection for Fraud Prevention in Project Monitoring. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi. 2025;16(4):1103-1111. doi:10.24012/dumf.1656802

BibTeX
@article{Sandkkaya2025,
  author = {Sandıkkaya, Mehmet Tahir and Tokdemir, Onur Behzat},
  title = {AI-Driven Anomaly Detection for Fraud Prevention in Project Monitoring},
  journal = {Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi},
  year = {2025},
  volume = {16},
  number = {4},
  pages = {1103--1111},
  doi = {10.24012/dumf.1656802}
}
Kostenlos starten →