Loan Approval & Risk Scoring System

Overview

• A machine learning project that predicts loan approval and estimates a loan risk score using applicant financial data
• Demonstrates real-world loan decision logic using classification and regression

Machine Learning Used

• Supervised Learning
• Logistic Regression for loan approval
• Linear Regression for risk score prediction

Dataset

• Loan application data containing income, loan amount, loan term, credit history, education, employment status, and property area
• Loan_ID column is removed as it is only an identifier

Workflow

• Load and clean the dataset
• Handle missing values and encode categorical variables
• Perform feature engineering
• Generate a rule-based loan risk score
• Train regression and classification models
• Evaluate model performance
• Accept user input and predict results

Output

• Loan risk score (0–100)
• Final loan decision:
  • Low Risk → Approved
  • Medium Risk → Approved with Conditions
  • High Risk → Rejected