This project aims to predict the selling price of used cars based on various features such as the car’s age, kilometers driven, fuel type, transmission, and number of previous owners. By using Machine Learning models, we can help car buyers and sellers make informed pricing decisions.

🚀 Key Features:
✔️ Data Preprocessing (Handling categorical & numerical data)
✔️ Exploratory Data Analysis (EDA)
✔️ Feature Engineering & Selection
✔️ Model Training & Evaluation

The dataset contains 301 entries with the following 9 features:

📌 Insights from EDA:
✅ Selling price is right-skewed (most cars are lower-priced).
✅ Present Price has the highest correlation with Selling Price.
✅ Fuel Type: Petrol cars dominate, followed by Diesel.
✅ Transmission Type: Manual cars are more common than automatic.

✔️ One-hot encoding for categorical features.
✔️ Feature scaling for numerical values.
✔️ Dropped irrelevant features like Car_Name.
✔️ Splitting dataset into 80% Training and 20% Testing.

# Splitting data into train and test sets
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

We experimented with different models:
✅ Linear Regression
✅ Random Forest Regressor
✅ Decision Tree
✅ XGBoost

📊 Performance Metrics Used:

• R² Score (How well the model fits the data)
• Mean Absolute Error (MAE)

📌 Best Model: XGBoost with 94% accuracy 🎯

pip install pandas numpy matplotlib seaborn scikit-learn xgboost

jupyter notebook

Open Car Price Prediction with Machine Learning.ipynb and run all cells.

🔹 Improve feature selection & engineering.
🔹 Try Deep Learning models.
🔹 Build a web app using Flask / Streamlit for real-time predictions.

This project successfully predicts used car prices with high accuracy using machine learning techniques. The XGBoost model provided the best results with a 94% R² Score.

💻 GitHub | 🌐 LinkedIn | ✉️ Email