Predictive system for pricing used vehicles in Lima based on RestNet and XGBOOST

Nunez Y.; Grados P.; Montero R. · International Conference on Engineering and Emerging Technologies, ICEET (2025) · DOI: 10.1109/iceet67911.2025.11424137

Resumen

The used car market faces multiple challenges in price prediction, especially in unstable economic environments, where factors such as supply, demand, and market fluctuations complicate the valuation process. This study proposes an advanced predictive system that integrates machine learning models, using ResNet for vehicle image analysis and XGBoost for processing structured data such as age, mileage, model, and other vehicle characteristics. The main objective of this research is to improve the accuracy of used car price prediction, while also considering economic variations that impact market valuation. By validating this model with specific data from the Peruvian market, the aim is to offer a robust tool that allows sellers and buyers to make more informed decisions. This innovative approach would contribute to a more transparent and objective evaluation of used vehicles, benefiting both buyers and sellers in the buying and selling process.

Instituciones: Universidad Peruana de Ciencias Aplicadas