What do we do?

Our project is aimed at improving the accuracy of fraud detection in banking transactions by using accurate machine learning models and innovative ML/AI weighted average systems.

Why this project?

One of our team members recently received this suspicious charge to their bank account. Luckily, it was marked as suspicious by the bank and the transaction never went through, but this got us thinking: how much fraud slips through the cracks? Thus, our mission statement and goal: to improve fraud detection through a more comprehensive algorithm.

What Makes Us Unique

Through the innovative use of a weighted averaging technique with a number of powerful machine learning algorithms, we are able to mitigate the weaknesses of each algorithm, and play to the strengths of the models. By combining k-nearest neighbors (KNN), logistic regression (LR), decision trees (DT), and random forest (RF), we are able to generate a robust and accurate bank transaction fraud detection model.

How It Works

We employ a blend of advanced machine learning models, offering customers
a highly effective tool for identifying fraudulent activities.

Our Models

  • KNNK-nearest neighbors is highly effective in both the complex relationships seen in fraud detection, as well as in adapting to changing patterns.
  • LRLogistic Regression works exceptionally well with linearly separable data and is highly interpretable, making it a staple in our toolkit.
  • DTDecision Tree is extremely simple and interpretable and generally handles both linear and non-linear data well.
  • RFRandom Forest is known for its high predictive accuracy and its ability to handle non-linear data effectively.

Use our program

⚠ Currently Under Development ⚠