Automobile Price Prediction Using Linear Regression

Project Overview

This project develops a machine learning regression model to predict automobile prices using vehicle specifications and technical characteristics.

The workflow includes:

• Data preprocessing
• Missing value handling
• Feature engineering
• Chi-Square feature selection
• Pearson correlation analysis
• Linear Regression modeling
• Model evaluation

---

Dataset Information

Dataset: Automobile Dataset

Target Variable:

price

Features include:

• make
• fuel-type
• aspiration
• body-style
• horsepower
• engine-size
• curb-weight
• city-mpg
• highway-mpg

---

Project Structure

automobile-price-prediction-linear-regression/
│
├── Automobile.csv
│
├── Automobile_Price_Prediction.ipynb
│
├── Images/
│   ├── price_distribution.png
│   ├── chi_square_feature_selection.png
│   ├── pearson_heatmap.png
│   ├── actual_vs_predicted.png
│   └── residual_plot.png
│
├── README.md
│
└── requirements.txt

---

Visualizations

Price Distribution

Chi-Square Feature Selection

Pearson Correlation Heatmap

Actual vs Predicted

Residual Plot

---

Model Performance

Evaluation Metrics:

• Mean Squared Error (MSE)
• Root Mean Squared Error (RMSE)
• Mean Absolute Error (MAE)
• Mean Absolute Percentage Error (MAPE)
• R² Score

Result

R² Score = 82.73%
MAPE = 21.82%

---

Technologies Used

• Python
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Scikit-Learn
• Jupyter Notebook

---