Hybrid process mining and fuzzy relation weighting for process violation weighting in fraud detection

Huda, Solichul and Shidik, Guruh Fajar and Rafrastara, Fauzi Adi and Abdollah, Mohd Faizal (2025) Hybrid process mining and fuzzy relation weighting for process violation weighting in fraud detection. International Journal of Intelligent Engineering and Systems, 18 (5). pp. 910-924. ISSN 2185-310X

Text 0103625112025935482558.pdf Download (532kB)

Abstract

There are many companies that implement Enterprise Resource Planning (ERP) to control their business processes. The risk of this implementation is the fact that fraud incidents in business processes also increase. Previous studies have proposed hybrid process mining with Fuzzy ARL and process mining with Heuristic Algorithm to detect fraud, but detection errors still occur because these methods cannot identify middle violations. This paper analyzes event logs in depth to determine employee relation weights during their activities. The relation weights obtained by hybrid with process mining are proposed to detect fraud. The proposed method integrates relational weights, process mining, and Fuzzy Multi-Attribute Decision Making to detect fraud. The relational weight method is used to determine the weight of the relation between employees. Process mining is used to compare the recorded event logs with the Standard Operating System (SOP). Finally, Fuzzy Multi-Attribute Decision Making is used to detect fraud. Using the same public dataset, the experimental results show that the process mining with Heuristic miner method obtained an accuracy of 0.9275, while process mining with the fuzzy ARL method obtained an accuracy of 0.9425, and process mining with the Relation weight obtained an accuracy of 0.96. Therefore, process mining and the Relation weight can detect fraud with medium violations and reduce false negatives.

Item Type:	Article
Uncontrolled Keywords:	Fraud detection, Process mining, Anomalies, Business process
Divisions:	Faculty of Information and Communication Technology
Depositing User:	Norfaradilla Idayu Ab. Ghafar
Date Deposited:	18 May 2026 00:57
Last Modified:	18 May 2026 00:57
URI:	http://eprints.utem.edu.my/id/eprint/29853
Statistic Details:	View Download Statistic

Actions (login required)

View Item