FraudGuard

FraudGuard

Inspiration

I was inspired to create FraudGuard because financial institutions, especially banks, are losing billions of dollars to fraud annually. Newly released Federal Trade Commission data shows that consumers reported losing more than $10 billion to fraud in 2023, a 14% increase over 2022 losses. It’s crucial to leverage AI to prevent fraud and provide a secure platform for customers, reducing financial risks.

What it does

FraudGuard is an AI-powered fraud detection system that uses advanced machine learning algorithms and data analysis to identify outliers and detect potential financial fraud in transactions.

How we built it

We built FraudGuard by employing advanced data analysis techniques and three machine learning models: Logistic Regression, Decision Trees, and MLP (Multi-Layer Perceptron). These models work together to detect anomalies and fraudulent activities in financial transactions.

Challenges we ran into

One of the main challenges was the lack of access to actual financial transaction data. With real-world data, we could better understand patterns of fraud and design more accurate algorithms for fraud detection.

Accomplishments that we're proud of

We are proud of successfully building a robust fraud detection model and, more importantly, of our team's collective learning journey. Some of my group members were new to machine learning, and seeing them develop skills and contribute to such a meaningful project was rewarding.

What we learned

We learned a great deal about AI, machine learning, and fraud detection processes. We also learned about the importance of real-world data for training models and how critical collaboration is when working on complex problems that have the potential to impact people's lives.

What's next for FraudGuard

Next, we aim to secure internships where we can work with industry professionals, improve our model further, and deploy it in real-world environments. By refining the algorithm with real data, we can enhance its accuracy and effectiveness in fraud prevention.