Price Prediction of Ford Cars Applying Multiple Machine Learning Methods

Yicheng Wang

2024

Abstract

The growing supply of used cars and lower prices compared with their counterparts make the used car market highly competitive. Such a dynamic and sophisticated market underscores the necessity for accurate price prediction, which is crucial for both buyers and sellers. Nowadays, the popularisation of online databases and advanced machine learning techniques have made price prediction based on machine learning models an obvious trend in the used car trading industry. This study delves into the performance evaluation of multiple machine learning methods for predicting used Ford car prices. Utilizing a comprehensive dataset from Kaggle, encompassing 17,966 entries with nine distinct features, the researcher employed a battery of regression models, including linear regression (LR), decision tree (DT), random forest (RF) and eXtreme Gradient Boosting (XGBoost). The approach involved rigorous feature engineering, model training and cross-validation evaluation, employing Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R^2) as key performance indicators. The results indicate that XGBoost and RF surpass traditional models in predictive accuracy, with XGBoost demonstrating the highest R^2 value of 0.9347. This study compares the performance of several widespread models and offers practical implications for stakeholders seeking to enhance transactional outcomes through data-driven pricing strategies

Paper Citation

in Harvard Style

Wang Y. (2024). Price Prediction of Ford Cars Applying Multiple Machine Learning Methods. In Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI; ISBN 978-989-758-726-9, SciTePress, pages 280-285. DOI: 10.5220/0013214800004568

in Bibtex Style

@conference{ecai24,
author={Yicheng Wang},
title={Price Prediction of Ford Cars Applying Multiple Machine Learning Methods},
booktitle={Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI},
year={2024},
pages={280-285},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013214800004568},
isbn={978-989-758-726-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI
TI - Price Prediction of Ford Cars Applying Multiple Machine Learning Methods
SN - 978-989-758-726-9
AU - Wang Y.
PY - 2024
SP - 280
EP - 285
DO - 10.5220/0013214800004568
PB - SciTePress