Machine Learning Model of Credit Classification

Our model predicts credit risk at relatively high accuracy and recall, help to prevent banks and companies from lending high risk borrower, and ensure business performance and economic resillience

Comment

Inspiration

Sustainability in the financial sector prevents another round of crisis, proved to be a precautionary measure against a reoccurrence, collateral debt-related financial crisis in 2008

What it does

This work is to predict the fluctuation and change of the dynamic of the market, and continue to predict sustainable financial methodology in the loan.

How I built it

we first review a data set about credit risk and use data analysis to predict credit risks. After that, we apply 3 classification techniques - Logistic Regression, Naive Bayes (NB) classification, and Tree classification. Then we improved the methods by adopting ensemble, bagging, and boosting

Challenges I ran into

Finding ways to improve accuracy and recall

Accomplishments that I'm proud of

Credit risk could be predicted at relatively high accuracy and recall

What I learned

Building ensemble, bagging and boosting

What's next for Machine Learning Model of Credit Classification

Further improve the accuracy and recall

Built With

Share this project:

Updates