Inspiration
We wanted to explore how banks used machine learning to predict fraud.
What it does
Trained on a dataset of around 30,000 instances, it uses k nearest neighbor (which we calculated was
How we built it
Scikit learn, pandas, numpy
Challenges we ran into
Monzo failing to be a good API, overfitting machine learning models
Accomplishments that we're proud of
actually getting a working demo before demo day
What we learned
What's next for Xxx X-treme Bank Fraud.exe xxX
Built With
- capital-one
- matplotlib
- monzo
- numpy
- pandas
- python
- scikitlearn
- tensorflow
Log in or sign up for Devpost to join the conversation.