Skip to content

LKEthridge/Supervised_Learning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

7afcb61 Β· Jan 20, 2025

History

3 Commits
Jan 20, 2025
Jan 20, 2025
Jan 20, 2025
Jan 20, 2025
Jan 20, 2025

Repository files navigation

Supervised_Learning

This was a Supervised Learning project for TripleTen. πŸ‘©πŸ½β€πŸ’»

This project developed a Random Forest Classifier to predict customer churn for Beta Bank, achieving an F1 score of 0.61 and a strong AUC-ROC score despite class imbalance. By targeting likely-to-leave customers, the model provides a tool for optimizing retention strategies and aligning predictions with actual churn trends. This approach offers Beta Bank a data-driven solution to reduce customer attrition and secure its future.

Skills Highlighted

πŸ‘€ Supervised Learning 🧼 Feature Prep including One-Hot, Label, and Ordinal Encoding βš–οΈ Feature Scaling & Class-Imbalance Handling πŸ€” Confusion Matrices, Precision, Recall, and F1 Score ↕️ Imbalanced Classification with Upsampling or Downsampling πŸͺ¨ ROC-Curve, PR Curve, True Positive Rate, and False Positive Rate πŸ’― Regression Metrics

Installation & Usage

  • This project uses pandas, numpy, train_test_split, DecisionTreeClassifier, RandomForestClassifier, LogisticRegression, f1_score, roc_auc_score, accuracy_score, matplotlib.pyplot, shuffle, and StandardScaler. It requires python 3.9.6.