Classification with Playground Series - S4E10
Kaggle Playground October 2024 - Loan Approval Prediction
Goal: The goal for this competition is to predict whether an applicant is approved for a loan.
Evaluation: Submissions are evaluated using area under the ROC curve using the predicted probabilities and the ground truth targets.
Learnings:
- Handle class imbalance with SMOTE() and scale_pos_weight parameters in algorithms
- Use predict_proba in Gradient Boosting
- Use Pipeline and Column Transformer
Table of Contents
1. Preparation
2. Exploratory Data Analysis (EDA)
3. Feature Engineering (SMOTE, Encoding Pipeline)
5. Machine Learning (Gradient Boosting)
6. LightAutoML
7. Evaluate and Ensemble
8. Final Prediction
9. Submission
Achievement: My ROC AUC score is 0.98.
* Check out more details in my Medium blog.
