🚗 Car Price Prediction using Machine Learning

A machine learning project that predicts the selling price of used cars based on features like present price, fuel type, transmission, mileage, and car age.

Built as part of my CodeAlpha Data Science Internship

---

📌 Project Overview

Detail	Description
Objective	Predict resale price of used cars
Dataset	301 entries, 9 features
Models Used	Linear Regression, Random Forest Regressor
Best Model	Random Forest (highest R² score)
Tools	Python, Pandas, Scikit-learn, Matplotlib, Seaborn

---

📂 Project Structure

├── Car_Price_Prediction.ipynb   # Main Jupyter Notebook
├── car data.csv                 # Dataset
├── requirements.txt             # Python dependencies
└── README.md                    # Project documentation

---

📊 Dataset Features

Feature	Description
Car_Name	Name of the car
Year	Year of purchase
Selling_Price	Price the owner wants to sell (Target)
Present_Price	Current ex-showroom price
Driven_kms	Kilometers driven
Fuel_Type	Petrol / Diesel / CNG
Selling_type	Dealer / Individual
Transmission	Manual / Automatic
Owner	Number of previous owners

---

🔧 Workflow

1. Data Loading — Read CSV using Pandas
2. Exploratory Data Analysis (EDA) — Visualize distributions, correlations, and relationships
3. Feature Engineering — Created Car_Age from Year
4. Data Preprocessing — Label encoded categorical variables (Fuel_Type, Selling_type, Transmission)
5. Model Training — Trained Linear Regression and Random Forest models
6. Model Evaluation — Compared using MAE, RMSE, and R² Score

---

📈 Results

Model	MAE	RMSE	R² Score
Linear Regression	~1.3	~2.0	~0.84
Random Forest	~0.9	~1.3	~0.96

Random Forest significantly outperforms Linear Regression for this dataset.

---

🔑 Key Findings

• Present Price is the strongest predictor of selling price
• Car Age has a strong negative correlation with selling price
• Diesel cars tend to hold value better than Petrol cars
• Random Forest captures non-linear relationships much better than Linear Regression

---

🚀 How to Run

1. Clone the repository

   git clone https://github.com/kinzaemannn/CodeAlpha-Car-Price-Prediction.git
   cd CodeAlpha-Car-Price-Prediction

2. Install dependencies

   pip install -r requirements.txt

3. Open the notebook

   jupyter notebook Car_Price_Prediction.ipynb

   Or upload to Kaggle and run directly.

---

🛠️ Technologies Used

• Python 3.x
• Pandas — Data manipulation
• NumPy — Numerical operations
• Matplotlib & Seaborn — Data visualization
• Scikit-learn — Machine learning models and evaluation

---

📝 License

This project is for educational purposes as part of the CodeAlpha Data Science Internship.

---

👤 Author

Kinza Eman
🔗 Kaggle Profile
🔗 GitHub