# 🚗 Car Price Data Science Project

## 📌 Project Overview
This project explores a Kaggle dataset of 16,000+ used and new cars in Australia.  
The goal is to clean, analyze, and build predictive models for **car prices**.

## 📊 Key Steps
1. Data Cleaning:
   - Converted `Price`, `Kilometres`, `FuelConsumption`, etc. into numeric values
   - Extracted `Engine Capacity` and `Cylinders`
   - Created new features like `CarAge`, `PricePerKm`, and `LuxuryBrand`
   
2. Exploratory Data Analysis (EDA):
   - Distribution of car prices
   - Price vs Kilometres scatter plot
   - Average prices by brand
   - Correlation heatmap

3. Feature Engineering:
   - Added `LuxuryBrand` flag
   - Created `PricePerKm`
   - Derived `CarAge`

4. Predictive Modeling:
   - Built regression models to predict car price
   - Evaluated with R² and MAE

5. Clustering (optional):
   - Used KMeans to group cars into clusters (e.g., economy, mid-range, luxury)

## 📈 Tools & Libraries
- Python
- Pandas, NumPy
- Matplotlib, Seaborn
- Scikit-learn

## 🚀 Results
- Cleaned dataset of **16,000+ cars**
- Clear insights into pricing trends
- Predictive model to estimate car price
- Visualizations to support analysis

---