Team 15

Inspiration

We were inspired by the many applications of machine learning already seen within the world of computer science.

What it does

Our model is trained with the train sheet provided and can make predictions on any test sheet, detecting fraud based on similar characteristics.

How we built it

Using python, scikit-learn, and several other helpful modules, we were able to build our model.

Challenges we ran into

The biggest challenge we ran into was balancing the time our cross validation search took in order to find ideal parameters for our model. This meant that in order to run the model within a reasonable amount of time, some sacrifice to accuracy had to be made.

Accomplishments that we're proud of

We're incredibly proud of the entire project and how far we've gotten. The model was fine tuned to work as efficiently as we could get it within the time-span.

What we learned

We've learned a significant amount about machine learning models as well as preprocessing of data

What's next for Team 15

We'd like to continue fine-tuning this model outside of the competition in order to make it even more efficient, as well as continuing looking for new opportunities with machine learning.