🏠 House Price Prediction (Linear Regression)

A complete beginner-friendly Machine Learning project built using the Ames Housing dataset.
This project predicts the sale price of houses using Linear Regression, along with data cleaning,
outlier handling, and One-Hot Encoding for categorical variables.

---

📊 Project Overview

This project walks through the full process of building a regression model step-by-step:

1. Data Loading & Inspection
2. Data Cleaning
   • Handling missing values
   • Removing outliers (GrLivArea > 4000 & SalePrice < 300000)
3. Exploratory Data Analysis (EDA)
   • Correlation heatmaps
   • Key relationships (GrLivArea, OverallQual, etc.)
4. Model Building
   • Linear Regression (Baseline + Improved)
   • One-Hot Encoding for categorical variables
5. Model Evaluation
   • R² Score
   • RMSE
   • Visualization of Actual vs Predicted values
   • Residuals distribution

---

🚀 Results

Metric	Value
R² Score	~0.89
RMSE	~23,000

The model performs quite well for a simple Linear Regression model!

---

🧠 Key Learnings

• Importance of outlier removal before fitting a linear model
• How categorical encoding (One-Hot) improves regression performance
• Basic workflow of a data science project, from EDA → modeling → evaluation

---

🛠️ Technologies Used

• Python 🐍
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-learn

---

📂 Files in Repository

House_Price_Prediction_Ishu_Final.ipynb # Final Jupyter notebook README.md # Project overview

---

---

⭐ If you like this project, give it a star on GitHub! ⭐

📈 Future Improvements

• Try advanced models like Ridge, Lasso, or RandomForest
• Hyperparameter tuning for better performance
• Deploy model using Streamlit

---

👨‍💻 Author

Ishu Singh
📧 Email: [email protected]
🌐 GitHub: https://github.com/ishuz-data-Git

---

⭐ If you like this project, give it a star on GitHub! ⭐