Download PDF

Hybrid Reasoning for Bank Fraud Detection Using RDF Ontology: Conditional Python and SPARQL

Khaoula El Ater, Ouayres Oumaima, Abderrahman Chekry
1Cadi Ayad University, UCA, Polydisciplinary Faculty of Safi, Department of Mathematics and Computer Science, Morocco.
2Cadi Ayad University, UCA, Graduate School of Technology of Safi, LAPSSII Laboratory, Morocco.
acheky@uca.aca.ma
How to cite: Khaoula El Ater, Ouayres Oumaima, Abderrahman Chekry, "Hybrid Reasoning for Bank Fraud Detection Using RDF Ontology, Conditional Python and SPARQL", Sciences Methods and Technologies International Journal (SciMeTech), Vol 2, Issue 2, p 7-13
Abstract
Detection of bank frauds continues to be a serious problem within the finance industry, which entails the loss of billions of euros per year. In this paper, an innovative framework is introduced which combines ontological modeling with the use of OWL, dynamic rule-based reasoning by applying conditional Python programming, and querying through patterns with SPARQL. The framework makes use of the Owlready2, and it has been implemented as part of the web interface created using Streamlit. It allows manual input or upload of transaction data from CSV files, automatic analysis, and visualization of results. In this research paper, we build on previous studies about fraud detection using ontology. We apply some basic concepts that include ISO 20022, SWIFT, and Customer Transaction Ontology (CTO). The main purpose of our model is to provide an effective and modular framework for fraud detection by incorporating ontological modeling and rules.
Keywords: Bank fraud, Ontology, SPARQL, Python, OWL

References

  1. Auger, B., Chergui, H., Chehade, Y., El Kadri, J., Abrouk, L., & Cabioch, N. (2022). Construction of an ontology in the financial domain for fraud detection. INFORSID 2022.
  2. Lamy, J.- B. (2017). Owlready2: Ontology- oriented programming in Python. Knowledge- Based Systems, 93, 57- 69. https://doi.org/10.1016/j.knosys.2015.12.024
  3. Streamlit. (n.d.). Streamlit documentation. Retrieved June 12, 2025, from https://docs.streamlit.io
  4. Financial Action Task Force. (2021). Guidance on the risk- based approach to combating money laundering. Retrieved June 13, 2025, from https://www.fatf- gafi.org/
  5. Antoniou, G., & van Harmelen, F. (2008). A semantic web primer (2nd ed.). MIT Press.
  6. Berners- Lee, T., Hendler, J., & Lassila, O. (2001). The semantic web. Scientific American, 284(5), 28- 37. https://doi.org/10.1038/scientificamerican0501- 28
  7. Brickley, D., & Guha, R. V. (2014). RDF Schema 1.1. W3C Recommendation. Retrieved June 14, 2025, from https://www.w3.org/TR/rdf- schema/
  8. W3C OWL Working Group. (2012). OWL 2 Web Ontology Language Document Overview (Second Edition). Retrieved June 14, 2025, from https://www.w3.org/TR/owl2- overview/
  9. Prud'hommeaux, E., & Seaborne, A. (2008). SPARQL query language for RDF. W3C Recommendation. Retrieved June 14, 2025, from https://www.w3.org/TR/rdf- sparql- query/
  10. Kumar, R., & Ravi, V. (2007). Fraud detection in credit card transactions using hybrid machine learning models. International Journal of Computer Applications, 21(1), 39- 45.
  11. Zanin M. & Papo D. 2016. Detection of fraud in banking transactions using complex network theory. Scientific Reports, 6, 34190. https://doi.org/10.1038/srep34190
  12. Purohit, H., & Purohit, R. (2018). A study on ontology- based intelligent fraud detection system. International Journal of Computer Applications, 180(46), 6- 11.
  13. Zhao, J., Song, Y., & Wang, H. (2013). Fraud detection in online systems using probabilistic ontologies. Expert Systems with Applications, 40(15), 5973- 5983.
  14. Protege Team. (2021). Protege Ontology Editor (Version 5.6.5) [Software]. Stanford Center for Biomedical Informatics Research. Retrieved June 15, 2025, from https://protege.stanford.edu/
  15. Turban, E., Sharda, R., Delen, D., & King, D. (2018). Business intelligence, analytics, and data science: A managerial perspective (4th ed.). Pearson.
SciMeTech · Vol 2, Issue 2 · p 7-13