Indian Used Car Price Prediction
Project Overview
The primary objective of this data science project is to predict the price of used cars in major Indian metro cities by analyzing a wide range of car features. These features include the car's manufacturer, model, variant, fuel type, color, kilometers driven, body style, transmission type, manufacture date, model year, CNG kit availability, ownership history, dealer details, and quality scores.
About the Dataset:
The "Indian IT Cities Used Car Dataset 2023" is a comprehensive collection of data that provides valuable insights into the used car market across major metro cities in India. This dataset encompasses details such as car models, variants, pricing, fuel types, dealer locations, warranty information, colors, kilometers driven, body styles, transmission types, ownership history, manufacture dates, model years, dealer names, CNG kit availability, and quality scores.
Data Dictionary:
| Column Name | Description |
|---|---|
| ID | Unique ID for each listing |
| Company | Name of the car manufacturer |
| Model | Name of the car model |
| Variant | Name of the car variant |
| Fuel Type | Fuel type of the car |
| Color | Color of the car |
| Kilometer | Number of kilometers driven by the car |
| Body Style | Body style of the car |
| Transmission Type | Transmission type of the car |
| Manufacture Date | Manufacture date of the car |
| Model Year | Model year of the car |
| CngKit | Whether the car has a CNG kit or not |
| Price | Price of the car |
| Owner Type | Number of previous owners of the car |
| Dealer State | State in which the car is being sold |
| Dealer Name | Name of the dealer selling the car |
| City | City in which the car is being sold |
| Warranty | Warranty offered by the dealer |
| Quality Score | Quality score of the car |
Conclusion
From the exploratory data analysis, several key insights have been derived regarding the Indian used car market. These insights include demand and price trends, the influence of factors such as fuel type and color, the impact of kilometer readings and body styles, geographical price variations, and the role of quality scores in determining car prices.
Machine learning models, specifically the Decision Tree Regressor and Random Forest Regressor, were utilized to predict car prices, with the Random Forest Regressor proving more effective. Feature importance analysis revealed that car age, body style, and car manufacturer are key factors affecting car prices.
This project offers valuable insights for both buyers and sellers in the Indian used car market, facilitating data-driven decision-making and price predictions.
View Notebook Download Dataset