Our paper “A Systematic Survey and Empirical Comparison of Hybrid Methods for Imbalanced Fraud Detection” is now available in AUT Journal of Mathematics and Computing.

Fraud detection faces a critical hurdle: severe class imbalance. Our work systematically explores hybrid frameworks that combine resampling techniques with machine learning to overcome this challenge.
🔑 Key Contributions:
✅ Comprehensive survey of hybrid imbalance learning methods
✅ Rigorous empirical study on auto insurance fraud data
✅ Open-source code for reproducibility and further research
✅ Evidence that resampling strategy profoundly impacts model performance

📄 Access the paper:
https://ajmc.aut.ac.ir/article_5913.html

DOI: 10.22060/AJMC.2025.24642.1446

#FraudDetection #ImbalancedLearning #MachineLearning #Resampling #Oversampling #DataScience #AIResearch #OpenScience #InsuranceFraud

PDF on Researchgate Link: https://lnkd.in/dpmx7skm