Project Information:
- Languages/Tools: Python
- Project Date: Sep 2024 - Dec 2024
- Google Colab: Click Here
Description:
Our analysis evaluated metrics like accuracy, recall (focusing on "Not Approved" cases), and F1 score across Logistic Regression, K-Nearest Neighbors, Support Vector Machine (SVM), Decision Tree, and Random Forest models. Key features influencing loan approval predictions include Credit History, Loan Amount, and Coapplicant Income.
Results:
The Logistic Regression and SVM models emerged as top performers, achieving high recall and balanced metrics. Logistic Regression is preferred for its simplicity and interpretability, making it effective for flagging high-risk loan applications for review.
Model Deployment: Deploy Logistic Regression for its practicality and interpretability, or SVM if the problem demands capturing non-linear relationships in the data.