Used Car Price Prediction and Classification Using Machine Learning Approaches

The used car market in Egypt faces challenges related to inflation and inconsistent seller pricing, often involving unrealistic prices set to attract potential buyers. This affects the evaluation of fair market values. To resolve this issue, an artificial intelligence system integrating image recognition and price prediction techniques is introduced. The system employs machine learning methods to accurately predict car prices, with the XGBoost model achieving a prediction accuracy of 97%. Additionally, the system utilizes a DenseNet201 image recognition model, achieving 94% accuracy in identifying car models from images. This integrated approach simplifies the valuation process for users, especially those unfamiliar with detailed vehicle specifications, reducing errors from manual input. The proposed system adapts dynamically to market changes, offering reliable and updated pricing forecasts. By delivering precise and accessible pricing information, the tool promotes transparency and fairness in the used car market.