Stop card fraud before it costs you.
FraudShield combines four machine-learning algorithms with fuzzy membership to spot fraudulent card transactions, even when fraud cases are rare in the data.
The problem
Card fraud keeps growing, and keeps changing.
Fraudsters constantly adapt, and fraud cases are a tiny fraction of all transactions. This imbalance makes fraud hard for standard models to learn, leading to missed fraud or too many false alarms.
lost to payment card fraud worldwide in 2023.
machine-learning algorithms combined: LR, LDA, Naive Bayes and XGBoost.
membership logic added to handle uncertain, imbalanced cases.
Sources: Nilson Report (2024) for global card fraud losses; method from the founding research paper (ICBATS, 2025).
How it works
How it works
A clear path from data to decision.
Ingest
Card transactions stream in from the payment system.
Score
ML models assess each transaction's fraud likelihood.
Refine
Fuzzy membership handles borderline, uncertain cases.
Act
Suspicious transactions are flagged, held or verified.
Product
What makes it different.
Every feature is grounded in the founding research.
Built for imbalance
Designed for data where fraud is rare.
Model ensemble
Logistic regression, LDA, Naive Bayes and XGBoost compared and combined.
Fuzzy logic
Handles uncertainty better than hard thresholds.
Adaptable
Retrains as fraud patterns evolve.
Who it's for
Who it's for.
Banks
Real-time card fraud screening.
Payment processors
Protect merchants and cardholders.
E-wallets & fintechs
Fraud control without blocking good customers.
The research
Built on peer-reviewed research.
- Intelligent fraud detection using machine learning and fuzzy membership.
- Addresses class imbalance in credit card fraud data.
- Uses LR, LDA, Naive Bayes and XGBoost on a public imbalanced credit card dataset (IEEE DataPort).
- Evaluated with accuracy and other metrics and compared with previous work.
Alheeti, K. M. A., et al. (2025). Fraud Detection by Machine Learning and Fuzzy Using Imbalance Credit Card… Dataset. ICBATS 2025, IEEE.
Team
The team behind it.
Founded in 2025 by a student team from two faculties.
Faculty of Administrative & Financial Sciences
Banking, risk and fraud operations.
Faculty of Science (Computer Science)
Machine learning and fuzzy systems.
FAQ
Questions we hear often.
Does it block legitimate customers?
It is designed to reduce false alarms by handling uncertain cases with fuzzy logic.
Can it run in real time?
Yes. Scoring is lightweight enough for transaction-time decisions.
Is this a finished commercial product?
It is a research-based student startup moving from validated prototype toward pilots with real users.
Can we partner or run a pilot?
Yes. We are looking for organisations willing to test the solution in real conditions and share feedback.
Bring FraudShield to your organisation.
We are onboarding pilot partners now. Read the research and get in touch.