Skip to main content
SpringerLink
Log in

Advancing Fraud Detection Systems Through Online Learning

  • Conference paper
  • First Online:
Machine Learning and Knowledge Discovery in Databases: Applied Data Science and Demo Track (ECML PKDD 2023)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 14174))

Abstract

The rapid increase in digital transactions has led to a consequential surge in financial fraud, requiring an automatic way of defending effectively from such a threat. The past few years experienced a rise in the design and use by financial institutions of different machine learning-based fraud detection systems. However, these solutions may suffer severe drawbacks if a malevolent adversary adapts their behavior over time, making the selection of the existing fraud detectors difficult. In this paper, we study the application of online learning techniques to respond effectively to adaptive attackers. More specifically, the proposed approach takes as input a set of classifiers employed for fraud detection tasks and selects, based on the performances experienced in the past, the one to apply to analyze the next transaction. The use of an online learning approach guarantees to keep at a pace the loss due to the adaptive behavior of the attacker over a given learning period. To validate our methodology, we perform an extensive experimental evaluation using real-world banking data augmented with distinct fraudulent campaigns based on real-world attackers’ models. Our results demonstrate that the proposed approach allows prompt updates to detection models as new patterns and behaviors are occurring, leading to a more robust and effective fraud detection system.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 79.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 99.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    In the current work, we are assuming the attacker has access to the financial institution historical data.

  2. 2.

    An alternative approach is the user-centric approach, which consists in learning the concept of fraud corresponding to each customer or groups of customers with similar spending patterns.

  3. 3.

    From a technical point of view, in our dataset the values corresponding to IP, SessionID, and IBAN have been provided as hashed values. However, the above-described operations can be performed even having the data in hash form.

  4. 4.

    The code corresponding to the above procedure and supplementary material is provided at [1].

  5. 5.

    We used as loss the Normalized Cost measure defined in [24, 36] due to the imbalance nature of the dataset. Moreover, we used a filter method based on Pearson’s correlation for feature selection and 3-fold cross-validation to estimate the Normalized Cost metric for the different models.

  6. 6.

    We provide the F1 score of the different models on different attack strategies in supplementary material at [1].

  7. 7.

    Note that the loss of the 30 base models has not been included in the figure for visualization purposes, but can be found in [1].

  8. 8.

    The weight changes over time of the MWU methods are reported in [1].

References

  1. Advancing fraud detection systems through online learning - supplementary materials and source code. https://github.com/necst/advancing_fds_code (2023)

  2. AlAhmadi, B.A., Axon, L., Martinovic, I.: 99% false positives: A qualitative study of SOC analysts’ perspectives on security alarms. In: Butler, K.R.B., Thomas, K. (eds.) 31st USENIX Security Symposium, USENIX Security 2022, Boston, MA, USA, August 10-12, 2022, pp. 2783–2800. USENIX Association (2022), https://www.usenix.org/conference/usenixsecurity22/presentation/alahmadi

  3. Aziz, M., Kaufmann, E., Riviere, M.K.: On multi-armed bandit designs for dose-finding clinical trials. J. Mach. Learn. Res. 22(1), 686–723 (2021)

    MathSciNet  MATH  Google Scholar 

  4. Bagga, S., Goyal, A., Gupta, N., Goyal, A.: Credit card fraud detection using pipeling and ensemble learning. Proc. Comput. Sci. 173, 104–112 (2020). https://doi.org/10.1016/j.procs.2020.06.014, https://www.sciencedirect.com/science/article/pii/S1877050920315167, international Conference on Smart Sustainable Intelligent Computing and Applications under ICITETM2020

  5. Bahnsen, A.C., Aouada, D., Stojanovic, A., Ottersten, B.E.: Feature engineering strategies for credit card fraud detection. Expert Syst. Appl. 51, 134–142 (2016). https://doi.org/10.1016/j.eswa.2015.12.030

  6. Bekirev, A.S., Klimov, V.V., Kuzin, M.V., Shchukin, B.A.: Payment card fraud detection using neural network committee and clustering. Optical Memory Neural Netw. 24(3), 193–200 (2015). https://doi.org/10.3103/S1060992X15030030

    Article  Google Scholar 

  7. Bernasconi, M., Martino, S., Vittori, E., Trovò, F., Restelli, M.: Dark-pool smart order routing: a combinatorial multi-armed bandit approach. In: Proceedings of the Third ACM International Conference on AI in Finance, pp. 352–360 (2022)

    Google Scholar 

  8. Bhattacharyya, S., Jha, S., Tharakunnel, K.K., Westland, J.C.: Data mining for credit card fraud: A comparative study. Decis. Support Syst. 50(3), 602–613 (2011). https://doi.org/10.1016/j.dss.2010.08.008

  9. Brause, R.W., Langsdorf, T.S., Hepp, H.: Neural data mining for credit card fraud detection. In: 11th IEEE International Conference on Tools with Artificial Intelligence, ICTAI ’99, Chicago, Illinois, USA, November 8-10, 1999, pp. 103–106. IEEE Computer Society (1999). https://doi.org/10.1109/TAI.1999.809773

  10. Carcillo, F., Borgne, Y.L., Caelen, O., Bontempi, G.: An assessment of streaming active learning strategies for real-life credit card fraud detection. In: 2017 IEEE International Conference on Data Science and Advanced Analytics, DSAA 2017, Tokyo, Japan, October 19-21, 2017, pp. 631–639. IEEE (2017). https://doi.org/10.1109/DSAA.2017.10

  11. Carminati, M., Baggio, A., Maggi, F., Spagnolini, U., Zanero, S.: Fraudbuster: Temporal analysis and detection of advanced financial frauds. In: Giuffrida, C., Bardin, S., Blanc, G. (eds.) Detection of Intrusions and Malware, and Vulnerability Assessment - 15th International Conference, DIMVA 2018, Saclay, France, June 28-29, 2018, Proceedings. Lecture Notes in Computer Science, vol. 10885, pp. 211–233. Springer (2018). https://doi.org/10.1007/978-3-319-93411-2_10

  12. Carminati, M., Caron, R., Maggi, F., Epifani, I., Zanero, S.: Banksealer: A decision support system for online banking fraud analysis and investigation. Comput. Secur. 53, 175–186 (2015). https://doi.org/10.1016/j.cose.2015.04.002

  13. Carminati, M., Polino, M., Continella, A., Lanzi, A., Maggi, F., Zanero, S.: Security evaluation of a banking fraud analysis system. ACM Trans. Priv. Secur. 21(3), 11:1–11:31 (2018). https://doi.org/10.1145/3178370

  14. Carminati, M., Santini, L., Polino, M., Zanero, S.: Evasion attacks against banking fraud detection systems. In: Egele, M., Bilge, L. (eds.) 23rd International Symposium on Research in Attacks, Intrusions and Defenses, RAID 2020, San Sebastian, Spain, October 14–15, 2020, pp. 285–300. USENIX Association (2020), https://www.usenix.org/conference/raid2020/presentation/carminati

  15. Cartella, F., Anunciação, O., Funabiki, Y., Yamaguchi, D., Akishita, T., Elshocht, O.: Adversarial attacks for tabular data: Application to fraud detection and imbalanced data. In: Espinoza, H., McDermid, J.A., Huang, X., Castillo-Effen, M., Chen, X.C., Hernández-Orallo, J., hÉigeartaigh, S.Ó., Mallah, R. (eds.) Proceedings of the Workshop on Artificial Intelligence Safety 2021 (SafeAI 2021) co-located with the Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI 2021), Virtual, February 8, 2021. CEUR Workshop Proceedings, vol. 2808. CEUR-WS.org (2021). https://ceur-ws.org/Vol-2808/Paper_4.pdf

  16. Cesa-Bianchi, N., Lugosi, G.: Prediction, learning, and games. Cambridge University Press (2006)

    Google Scholar 

  17. Continella, A., Carminati, M., Polino, M., Lanzi, A., Zanero, S., Maggi, F.: Prometheus: Analyzing webinject-based information stealers. J. Comput. Secur. 25(2), 117–137 (2017). https://doi.org/10.3233/JCS-15773

  18. European Central Bank: Seventh report on card fraud. Tech. rep. (2021). http://web.archive.org/web/20230521043629/https://www.ecb.europa.eu/pub/cardfraud/html/ecb.cardfraudreport202110~cac4c418e8.en.html

  19. Forough, J., Momtazi, S.: Ensemble of deep sequential models for credit card fraud detection. Appl. Soft Comput. 99, 106883 (2021). https://doi.org/10.1016/j.asoc.2020.106883

  20. Freund, Y., Schapire, R.E.: Adaptive game playing using multiplicative weights. Games Econom. Behav. 29(1–2), 79–103 (1999)

    Article  MathSciNet  MATH  Google Scholar 

  21. Itoo, F., Singh, S., et al.: Comparison and analysis of logistic regression, naïve bayes and knn machine learning algorithms for credit card fraud detection. Int. J. Inf. Technol. 13(4), 1503–1511 (2021)

    Google Scholar 

  22. Kim, E., et al.: Champion-challenger analysis for credit card fraud detection: Hybrid ensemble and deep learning. Expert Syst. Appl. 128, 214–224 (2019). https://doi.org/10.1016/j.eswa.2019.03.042

  23. KPMG: Global Banking Fraud Survey. (2019). https://assets.kpmg/content/dam/kpmg/xx/pdf/2019/05/global-banking-fraud-survey.pdf

  24. Labanca, D., Primerano, L., Markland-Montgomery, M., Polino, M., Carminati, M., Zanero, S.: Amaretto: An active learning framework for money laundering detection. IEEE Access 10, 41720–41739 (2022). https://doi.org/10.1109/ACCESS.2022.3167699

  25. Marchal, S., Szyller, S.: Detecting organized ecommerce fraud using scalable categorical clustering. In: Balenson, D. (ed.) Proceedings of the 35th Annual Computer Security Applications Conference, ACSAC 2019, San Juan, PR, USA, December 09-13, 2019. pp. 215–228. ACM (2019). https://doi.org/10.1145/3359789.3359810

  26. Ministero dell’Economia e della Finanza: Rapporto statistico sulle frodi con le carte di pagamento. Tech. rep. (2021). https://www.dt.mef.gov.it/export/sites/sitodt/modules/documenti_it/antifrode_mezzi_pagamento/antifrode_mezzi_pagamento/Rapporto-statistico-sulle-frodi-con-le-carte-di-pagamento-edizione-2021.pdf

  27. Nuara, A., Trovò, F., Gatti, N., Restelli, M.: Online joint bid/daily budget optimization of internet advertising campaigns. Artif. Intell. 305, 103663 (2022)

    Article  MathSciNet  MATH  Google Scholar 

  28. Patidar, R., Sharma, L., et al.: Credit card fraud detection using neural network. Int. J. Soft Comput. Eng. (IJSCE) 1, 32–38 (2011)

    Google Scholar 

  29. Pozzolo, A.D., Boracchi, G., Caelen, O., Alippi, C., Bontempi, G.: Credit card fraud detection and concept-drift adaptation with delayed supervised information. In: 2015 International Joint Conference on Neural Networks, IJCNN 2015, Killarney, Ireland, July 12-17, 2015. pp. 1–8. IEEE (2015). https://doi.org/10.1109/IJCNN.2015.7280527

  30. Pozzolo, A.D., Boracchi, G., Caelen, O., Alippi, C., Bontempi, G.: Credit card fraud detection: A realistic modeling and a novel learning strategy. IEEE Trans. Neural Networks Learn. Syst. 29(8), 3784–3797 (2018). https://doi.org/10.1109/TNNLS.2017.2736643

  31. Randhawa, K., Loo, C.K., Seera, M., Lim, C.P., Nandi, A.K.: Credit card fraud detection using adaboost and majority voting. IEEE Access 6, 14277–14284 (2018). https://doi.org/10.1109/ACCESS.2018.2806420

  32. Sahin, Y., Duman, E.: Detecting credit card fraud by decision trees and support vector machines. In: World Congress on Engineering 2012. July 4–6, 2012. London, UK. vol. 2188, pp. 442–447. International Association of Engineers (2010)

    Google Scholar 

  33. Seeja, K., Zareapoor, M.: Fraudminer: A novel credit card fraud detection model based on frequent itemset mining. Sci. World J. 2014, 252797 (2014)

    Google Scholar 

  34. Sohony, I., Pratap, R., Nambiar, U.: Ensemble learning for credit card fraud detection. In: Ranu, S., Ganguly, N., Ramakrishnan, R., Sarawagi, S., Roy, S. (eds.) Proceedings of the ACM India Joint International Conference on Data Science and Management of Data, COMAD/CODS 2018, Goa, India, January 11-13, 2018, pp. 289–294. ACM (2018). https://doi.org/10.1145/3152494.3156815

  35. Trovò, F., Paladino, S., Restelli, M., Gatti, N.: Improving multi-armed bandit algorithms in online pricing settings. Int. J. Approx. Reason. 98, 196–235 (2018)

    Article  MathSciNet  MATH  Google Scholar 

  36. Whitrow, C., Hand, D.J., Juszczak, P., Weston, D.J., Adams, N.M.: Transaction aggregation as a strategy for credit card fraud detection. Data Min. Knowl. Discov. 18(1), 30–55 (2009). https://doi.org/10.1007/s10618-008-0116-z

  37. Xuan, S., Liu, G., Li, Z., Zheng, L., Wang, S., Jiang, C.: Random forest for credit card fraud detection. In: 15th IEEE International Conference on Networking, Sensing and Control, ICNSC 2018, Zhuhai, China, March 27-29, 2018, pp. 1–6. IEEE (2018). https://doi.org/10.1109/ICNSC.2018.8361343

  38. Zhang, Y., Tong, J., Wang, Z., Gao, F.: Customer transaction fraud detection using xgboost model. In: 2020 International Conference on Computer Engineering and Application (ICCEA), pp. 554–558 (2020). https://doi.org/10.1109/ICCEA50009.2020.00122

Download references

Acknowledgements

This paper is supported by the FAIR (Future Artificial Intelligence Research) project, funded by the NextGenerationEU program within the PNRR-PE-AI scheme (M4C2, Investment 1.3, Line on Artificial Intelligence).

Author information

Authors and Affiliations

  1. DEIB, Politecnico di Milano, Milan, Italy

    Tommaso Paladini, Martino Bernasconi de Luca, Michele Carminati, Mario Polino, Francesco Trovò & Stefano Zanero

Authors
  1. Tommaso Paladini
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Martino Bernasconi de Luca
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Michele Carminati
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Mario Polino
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Francesco Trovò
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Stefano Zanero
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Michele Carminati .

Editor information

Editors and Affiliations

  1. CENTAI, Turin, Italy

    Gianmarco De Francisci Morales

  2. NYU and Two Sigma, New York, NY, USA

    Claudia Perlich

  3. Netflix, Los Angeles, CA, USA

    Natali Ruchansky

  4. Telefonica Research, Barcelona, Spain

    Nicolas Kourtellis

  5. Politecnico di Torino, Turin, Italy

    Elena Baralis

  6. CENTAI, Turin, Italy

    Francesco Bonchi

Ethics declarations

Ethical Issues

Machine learning models have become increasingly ubiquitous in decision-making processes across various industries, especially financial fraud detection ones. However, the ethical implications of these models have come under scrutiny due to the potential for bias. Focusing on our work, if the base models are biased, any approach built upon them may also be biased. This is especially concerning when the models are used in sensitive areas such as fraud detection. On the other hand, since we do not explicitly exploit transaction features, we may not introduce further bias directly. However, it is important to note that the data used to train the models may still contain hidden biases that could influence the model’s predictions. Therefore, it is essential to ensure that the data sets used to train the models are diverse and representative of the population to minimize bias and prevent harm to vulnerable groups, as stated by the guidelines by the EU on AI methods (https://artificialintelligenceact.eu/).

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Paladini, T., Bernasconi de Luca, M., Carminati, M., Polino, M., Trovò, F., Zanero, S. (2023). Advancing Fraud Detection Systems Through Online Learning. In: De Francisci Morales, G., Perlich, C., Ruchansky, N., Kourtellis, N., Baralis, E., Bonchi, F. (eds) Machine Learning and Knowledge Discovery in Databases: Applied Data Science and Demo Track. ECML PKDD 2023. Lecture Notes in Computer Science(), vol 14174. Springer, Cham. https://doi.org/10.1007/978-3-031-43427-3_17

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-43427-3_17

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-43426-6

  • Online ISBN: 978-3-031-43427-3

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Societies and partnerships

Search

Navigation