classification-personal-loan

This research uses the Bank Personal Loan Modeling dataset, which aims to predict whether a person will approve a personal loan product from a bank based on customer demographic and behavioral features. The dataset used comes from the : https://www.kaggle.com/datasets/krantiswalke/bank-personal-loan-modelling. The dataset is analyzed using Support Vector Machine (SVM) algorithm with RBF kernel to handle binary classification, where the prediction focuses on the Personal Loan target variable (approved or not).

goals :

1. Helps banks identify customers who have the potential to approve loans.
2. Reduce the possibility of losses due to off-target lending.
3. Focuses marketing campaigns on customers who are more likely to approve loans.
4. Using SMOTE to handle class imbalance and improve prediction accuracy.
5. Speed up the loan approval process with more accurate decisions.

insights :

1. With accurate predictive models, banks can be more precise in determining which customers are eligible to receive loan offers.
2. Ensure that loans are only given to customers who actually have the potential to approve, thus reducing the risk of bad debts.
3. Banks can allocate marketing budgets more efficiently by targeting customers who have a higher probability of accepting loans.

Advices :

1. Add features related to customer behavior, such as payment history, to improve predictions.
2. Consider the customer's ability to repay the loan, not just approval.
3. Adjust loan terms for customers who are predicted to be more likely to approve.
4. Adjust loan terms for customers who are predicted to be more likely to approve.

Thank you for taking part in this research. For feedback, further questions, or deeper discussion, please feel free to contact me via email at dimasariyanto830@gmail.com or via LinkedIn at https://www.linkedin.com/in/dimas-ariyanto-a72038327/.

python #personalloan #classification #smote