Loan Defaulter Prediction using Machine Learning

This project predicts loan defaulters borrowers likely to default using historical applicant and loan information.
By analysing past financial and demographic data, the model identifies high-risk borrowers before approval, helping lenders reduce credit losses.

---

📊 Problem Statement

• Objective: Detect borrowers at risk of default before loan approval.
• Positive class: Defaulter (minority class in the dataset).
• Business goal: Maximise Recall on defaulters (catch as many risky borrowers as possible) while keeping Precision acceptable.
• Challenge: The dataset is highly imbalanced (few defaulters vs many non-defaulters), which can bias models toward predicting only non-defaulters.

---

🛠 Tech Stack

• Language: Python (Jupyter Notebook)
• Libraries: Pandas, NumPy, Matplotlib, Seaborn, Scikit-learn, XGBoost, imbalanced-learn
• Tools: Google Colab, GitHub

---