🏦 Credit Risk Prediction using Machine Learning

This project applies machine learning to assess the risk of granting credit to potential borrowers. The goal is to predict whether a loan applicant is likely to default, helping lenders make data-driven decisions.

---

🔍 Problem Statement

Credit risk modeling is crucial for financial institutions to minimize losses. This project uses historical customer data to predict loan status (safe or risky), leveraging classification techniques.

---

📊 Dataset Overview

The dataset includes key features such as:

• Age, Job, Housing, Loan status
• Credit amount, Duration, Purpose
• Savings, Checking account balance
• Personal status, Employment, and more

Source: Public dataset (e.g., UCI German Credit — replace if different)

---

🧠 ML Techniques Used

• EDA & Visualization (missing value treatment, correlations)
• Encoding & Scaling (Label Encoding, StandardScaler)
• Feature Engineering (selecting top features)
• Modeling: Logistic Regression, Random Forest, Decision Tree
• Evaluation: Accuracy, Precision, Recall, F1-Score, Confusion Matrix
• ROC Curve & AUC Score

---

✅ Results

• Achieved over 85% accuracy with Random Forest Classifier.
• Visualized results using Seaborn and Matplotlib.
• Feature importance and confusion matrix help interpret performance.

---

📌 Key Learnings

• End-to-end pipeline of a real ML classification problem.
• Importance of balancing data and choosing the right metric.
• Clear understanding of model performance and tradeoffs.

---