RawanIhab103/Egyptian_Used_Car_Price_Prediction
Record type:
modelproject
Creator:
Raw
Host:
A machine learning model predicting used car prices in Egypt using Scikit-Learn, with data preprocessing using Pandas and NumPy.
# Egyptian Used Car Price Prediction 🚗🇪🇬
## 📌 Project Overview
This project is an end-to-end Machine Learning regression model developed as part of the **NTI Machine Learning Training Program**. The goal is to estimate a fair market price for used cars in the Egyptian market based on various features.
Buyers and sellers often struggle to price cars accurately, leading to inconsistent estimations. This model solves the problem by providing a data-driven approach to predict realistic asking prices.
## 📊 Dataset Features
The model predicts the car price using the following attributes:
* Brand
* Model Year
* Mileage (Km)
* Fuel Type
* Transmission
* Color
* Location
## 🛠️ Tools & Technologies Used
* **Data Manipulation & Preprocessing:** Pandas, NumPy
* **Machine Learning Algorithms:** Scikit-Learn (Regression Models)
* **Development Environment:** Google Colab
## 📈 Model Evaluation
Several regression algorithms were trained and evaluated to find the best fit for predicting used car prices. The performance was measured using Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R-squared (R2) metrics.
| Model | MAE | MSE | RMSE | R2 Score |
| :--- | :--- | :--- | :--- | :--- |
| Decision Tree | 110,269 | 78,521,195,010 | 280,216 | 0.6845 |
| Random Forest | 86,710 | 43,503,644,516 | 208,575 | 0.8252 |
| **Gradient Boosting** | **83,622** | **35,016,490,037** | **187,127** | **0.8593** |
| HistGradientBoosting | 92,473 | 40,292,253,453 | 200,729 | 0.8381 |
| XGBoost | 86,431 | 38,599,853,874 | 196,468 | 0.8449 |
🏆 **Conclusion:** The **Gradient Boosting** model outperformed the others, achieving the highest variance explanation with an **R2 score of ~85.9%** and the lowest error rates. This makes it the most reliable model for estimating car prices in this dataset.
## 🚀 How to View and Run
1. The entire workflow, from data cleaning to model evaluation, is documented in the `project_final11.ipynb` notebook.
2. You can view the co …