Scholarlink

The exponential growth observed in the used car market has created opportunities for algorithms that can predict prices accurately. This study seeks to meet this goal by augmenting classical machine learning approaches with two powerful ensemble methods: Bagging and Boosting. The analysis improves the prediction capability and robustness of the model by also including some important features which are regards as secondary such as warranty and insurance. Various models including Random forest (bagging), XGBoost (boosting) among others were utilized and performance evaluated through R squared (R2), Mean Absolute Error (MAE) as well as Mean squared error (MSE). The proposed Random forest model gained an R 2 score of 0.95 which shows that it performed best for capturing complex feature-to-feature interactions and minimizing errors. The fact that this work delivers a viable and valid framework to use in estimating prices therefore aids buyers as well as sellers in the used car market. The findings emphasize the necessity of dataset enrichment and improvement in predictive algorithms in dealing with market intricacies and enhancing accountability.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhanced Used Car Price Prediction Using Ensemble Machine Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Saibaba Velidi,
  • Naga Charan Vaartha,
  • Koduri Varun Venkata Naga Sai Gupta,
  • Shaik Mohammed Haneef,
  • Sujith Kumar Bheemarasetty

摘要

The exponential growth observed in the used car market has created opportunities for algorithms that can predict prices accurately. This study seeks to meet this goal by augmenting classical machine learning approaches with two powerful ensemble methods: Bagging and Boosting. The analysis improves the prediction capability and robustness of the model by also including some important features which are regards as secondary such as warranty and insurance. Various models including Random forest (bagging), XGBoost (boosting) among others were utilized and performance evaluated through R squared (R2), Mean Absolute Error (MAE) as well as Mean squared error (MSE). The proposed Random forest model gained an R 2 score of 0.95 which shows that it performed best for capturing complex feature-to-feature interactions and minimizing errors. The fact that this work delivers a viable and valid framework to use in estimating prices therefore aids buyers as well as sellers in the used car market. The findings emphasize the necessity of dataset enrichment and improvement in predictive algorithms in dealing with market intricacies and enhancing accountability.