📊💸 Loan Data Analysis Project

Welcome to the Loan Data Analysis project! This repository contains an exploratory data analysis (EDA) notebook focused on identifying patterns and trends in loan applications.

---

📁 Dataset Overview

The dataset contains the following key features:

• 🆔 Loan_ID
• 👨‍💼 Gender
• 💍 Married
• 👶 Dependents
• 🎓 Education
• 💼 Self_Employed
• 💰 ApplicantIncome
• 🤝 CoapplicantIncome
• 🏦 LoanAmount
• ⏳ Loan_Amount_Term
• 🧾 Credit_History
• 🌐 Property_Area
• ✅ Loan_Status

---

🎯 Key Objectives

• 🔍 Perform exploratory data analysis (EDA)
• 📈 Understand relationships between features
• 🧠 Identify key factors influencing loan approval

---

🧰 Tools & Libraries Used

• 🐍 Python
• 🧮 NumPy
• 🐼 Pandas
• 📊 Matplotlib
• 🌈 Seaborn

---

🧼 Data Cleaning

• ✅ Handled missing values
• 🔄 Encoded categorical variables
• 🧹 Removed outliers using boxplots and replaced them with mean values

---

📊 Analysis Highlights

🔹 Univariate Analysis

• Used histograms, countplots, and pie charts to analyze distributions
• Found most applicants are male, married, and graduates

🔹 Bivariate Analysis

• Explored relationships between:
  • 💳 Credit_History & Loan_Status
  • 🧑‍🎓 Education & LoanAmount
  • 🏘️ Property_Area & Loan_Status

🔹 Income Insights

• 💼 Higher Applicant Income usually means larger loans
• 👥 Coapplicant Income has less influence on loan size

🔹 Outlier Detection

• Used box plots to identify outliers in ApplicantIncome and LoanAmount
• Replaced extreme outliers with mean values to normalize the data

---

📈 Visualizations

• 📦 Boxplots to show income distribution and outliers
• 🧊 Correlation Heatmap to show numeric relationships
• 📉 Scatter plots & Pair plots to reveal hidden patterns
• 🟩 Stacked bar charts to visualize class breakdowns by groups

---

🧠 Key Insights

1. ✅ Credit History is the most important factor for loan approval
2. 🎓 Graduates are slightly more likely to get loans approved
3. 🧍‍♂️ Male applicants dominate the dataset
4. 🏠 Semiurban property areas have higher approval rates
5. 📊 Loan amount and income are positively correlated

---

🔮 Future Work

• 🤖 Train machine learning models for loan status prediction
• 🧱 Feature engineering for improved performance
• 🧪 Hyperparameter tuning and model evaluation

---

Suggestion In this project I have analyse that we have to focused on semiurban areas and graduates people for better loan approval .

🏁 How to Run

1. Clone the repo:

   git clone https://github.com/AADARSH028/loan-data-analysis.git