🏠 House Price Prediction

A machine learning regression project that predicts residential house prices based on property features such as area, number of bathrooms, stories, and amenities — and identifies which features influence price the most.

📌 Built as part of the Xylofy Internship — Week 1 Project

---

📖 Overview

Real estate buyers and sellers often rely on guesswork or outdated comparisons to estimate a property's fair value. This project builds a regression model that predicts house prices from property features, evaluates model accuracy, and surfaces actionable insights for real estate businesses — backed by data, not intuition.

---

📊 Dataset

• Source: Kaggle — Housing Prices Dataset
• Records: 545 houses
• Features: 13 (area, bedrooms, bathrooms, stories, mainroad, guestroom, basement, hotwaterheating, airconditioning, parking, prefarea, furnishingstatus, price)
• Target variable: price (in INR)
• Missing values: None

---

🛠️ Tech Stack

Tool	Purpose
Python 3.x	Core programming language
Jupyter Notebook	Development environment
Pandas	Data loading & cleaning
Scikit-learn	Model training & evaluation
Matplotlib / Seaborn	Data visualization
ipywidgets	Interactive price predictor

---

⚙️ Project Workflow

1. Data Loading & Exploration

• Loaded the dataset with Pandas
• Explored shape, column types, and missing values
• Identified price as the target variable

2. Data Cleaning

• Verified no missing values or duplicates
• Converted yes/no columns (mainroad, guestroom, basement, hotwaterheating, airconditioning, prefarea) to binary (1/0)
• One-hot encoded furnishingstatus

3. Model Building

• Split data 80/20 into train and test sets
• Trained and evaluated three regression models:

Model	MAE	RMSE	R² Score
Linear Regression	9,70,043	13,24,506	0.65
Random Forest	10,22,560	14,01,496	0.61
Decision Tree	—	—	—

Linear Regression performed best, explaining ~65% of the variance in house prices.

4. Visualizations

• 📊 Price distribution histogram
• 🔥 Correlation heatmap
• 🎯 Actual vs. predicted price scatter plot
• 📈 Price vs. area (colored by bedrooms)
• 🌟 Feature importance bar chart
• 🥧 Market segmentation pie chart (Budget / Mid-Range / Luxury)

5. Insights & Recommendations

• Top price drivers: area, bathrooms, stories, air conditioning, main road access
• Surprising finding: number of bedrooms had less impact on price than expected
• Business recommendation: highlight area, bathrooms, and AC in listings; preferred-area properties command the highest premiums

---

🤖 Bonus: Interactive Price Predictor

An interactive widget built with ipywidgets lets you estimate a house price using sliders and dropdowns — no code editing needed.

Features:

• Real-time price prediction
• Price category (Budget / Mid-Range / Luxury)
• Price per sq.ft.
• ±10% confidence range
• Personalized tips to increase property value

# Example: Run the widget cells in order, then move the sliders
# and click "🏠 Predict Price"

⚠️ Note: Run all cells in order (Run → Run All) before using the widget — sliders and the prediction button depend on earlier cells being executed first.

---

📁 Repository Structure

HousePricePrediction_RitikaSen/
│
├── analysis.ipynb          # Complete notebook with all tasks
├── Housing.csv             # Dataset used
├── summary.pdf             # 1-page written summary of findings
├── charts/                 # All visualizations as PNG
│   ├── chart1_price_distribution.png
│   ├── chart2_correlation_heatmap.png
│   ├── chart3_actual_vs_predicted.png
│   ├── chart4_price_vs_area.png
│   ├── chart5_feature_importance.png
│   └── chart6_market_segmentation.png
└── README.md

---

🚀 How to Run

1. Clone the repository

   git clone https://github.com/<your-username>/HousePricePrediction_RitikaSen.git
   cd HousePricePrediction_RitikaSen

2. Install dependencies

   pip install pandas numpy matplotlib seaborn scikit-learn ipywidgets

3. Launch Jupyter Notebook

   jupyter notebook

4. Open analysis.ipynb and run all cells

   Run → Run All Cells
   

---

👤 Author

Ritika Sen Xylofy Internship — Week 1 Project | June 2026

---

📄 License

This project is for educational purposes as part of an internship assignment.