In this project, I analyzed customer's historical financial data and predicted if he/she will repay the loan or not
To predict if a new customer applying for a loan will have a good credit rating or a bad credit rating
Simply put, credit rating is the ability of a person to fulfil their financial commitments. This rating is arrived at, using the historical financial data of the person.
Since it is a classification problem, in this project I used K-Nearest Neighbors
- Algorithm used: K-Nearest Neighbor
- Optimal K-value selection: Based on accuracy scores for various K values, k-fold cross validation
- Validation techniques: Accuracy score, confusion matrix, ROC curve, AUC score
- Python
- data exploration/descriptive statistics
- data processing/cleaning
- Data Visualization
- Predictive Modeling