Loan-Default-Risk-Management-System-

🏦 Loan Default Risk Management System

A full-stack predictive analytics solution designed to assess loan default risk across diverse customer segments. This project analyzes 1,000,000+ loan applications to empower financial institutions with data-driven insights, risk modeling, and interactive dashboards for smarter credit decisioning and portfolio management.

🚗 GitHub Project Repository

🔗 Click to view Loan-Default-Risk-Management-System

🧠 Project Overview

Loan defaults pose a significant challenge to financial stability and lending efficiency. This project delivers an end-to-end analytics platform that enables:

🔍 Risk profiling of loan applicants
📊 Portfolio-level default trend analysis
🧠 Predictive modeling for credit decisions
📈 Dashboard-driven insights for compliance and strategy

🎯 Key Objectives

Clean and preprocess large-scale loan application data
Engineer features for default prediction and dashboarding
Build classification models to assess loan risk
Deploy interactive dashboards for stakeholder decision-making

📁 Project Structure

File Name	Description
`loan Default risk managemnt cleaned.csv`	Alternate cleaned dataset version
`cleaned_loandefault.csv`	Preprocessed dataset with feature engineering
`loan_risk_model.pkl`	Trained model for predicting loan default
`model_features.pkl`	Feature list used in model training
`loandefault.sql`	SQL queries for data extraction and filtering
`sqlconnect.py`	Python script for SQL database connection
`app.py`	Streamlit app for dashboard deployment
`finance_dashboard.py`	Additional dashboard for financial KPIs
`test_request.py`	Script for model testing and API simulation
`Loan Default Risk Management System.ipynb`	Jupyter notebook with EDA, modeling, and insights

🧹 Data Preprocessing

Converted Application_Date to datetime format
Calculated Debt_to_Income and Income_Loan_Ratio
Imputed missing values in Property_Ownership
One-hot encoded categorical features (Employment_Status, Region, Loan_Purpose)
Removed outliers and normalized financial metrics

📈 Exploratory Data Analysis

📊 Default trends by region, credit score, and loan purpose
💰 Income vs. loan amount correlation
🏦 Approval channel impact on default rates
📅 Distribution of monthly installments and loan terms

🤖 Modeling Approach

Target Variable: Defaulted
Algorithms Used: Logistic Regression, Random Forest, XGBoost
Evaluation Metrics: Accuracy, Precision, Recall, F1 Score, ROC-AUC
Top Features: Credit_Score, Debt_to_Income, Past_Defaults, Loan_Amount