KopaPap

• 

  Logo

• 

  KopaPap interface

• 

  displaying the result

Inspiration

The need for instant credit scoring for an individual/ institution (SME) in Kenya (based on Machine learning) and give lower interest rates on loans and overdrafts as compared to digital lenders. Some banks and Saccos in Kenya that use traditional credit scoring techniques which is time consuming hence most people opt for more expensive but instant loans of the digital lenders

What it does

The machine learning algorithm predicts the probability that the applicant will pay back a loan based on a pre-trained model.

How I built it

The API is built using python programming language, XGB classifier algorithm and Flask

Challenges I ran into

The biggest challenge I had was in acquiring the data I used to train the model. Highly imbalanced data that lead to a highly biased model

Accomplishments that I'm proud of

I am proud that I was able to build the app despite the lack of data

What I learned

To be patient How to work with highly imbalanced data

What's next for KopaPap

I hope to expand this API to other East African countries within the next two years, and to other African countries within the next 5 years

Built With

• flask
• python
• xgbclassifier