Novel Multi-Classifier Evidence Ensemble Algorithm for Predicting Traveler Repurchases

Strategic Insights -

A recent study in the Journal of Mathematics has proposed a novel multi-classifier evidence ensemble algorithm for predicting traveler repurchases in the airline industry. According to the research, the algorithm integrates evidence theory with machine learning to account for uncertain, incomplete, and ambiguous data. The study was conducted by a team of researchers from the Harbin Institute of Technology, with financial support from the National Natural Science Foundation of China. The research concluded that the proposed algorithm outperforms traditional prediction models in terms of accuracy, precision, and recall, achieving over 80% accuracy in binary classification tasks.

Key Takeaways:

  • The multi-classifier evidence ensemble algorithm integrates evidence theory with machine learning to predict traveler repurchases in the airline industry, addressing the challenges of uncertain, incomplete, and ambiguous data.
  • The algorithm was trained using 29 behavioral features derived from a low-cost Chinese airline and achieved over 80% accuracy and precision in binary classification tasks.
  • Ablation experiments using four classifier combinations at different sampling rates (30%, 50%, and 70%) validated the robustness and effectiveness of the framework.
  • The proposed algorithm outperformed traditional prediction models in terms of overall predictive performance for analyzing airline passenger behavior.
  • The research was conducted by a team of researchers from the Harbin Institute of Technology, led by Luning Liu, with financial support from the National Natural Science Foundation of China.
  • The research has significant implications for the airline industry, where predicting traveler repurchases is a critical aspect of marketing strategy.

Statistics:

  • The algorithm achieved over 80% accuracy and precision in binary classification tasks.
  • The ablation experiments used four classifier combinations at different sampling rates (30%, 50%, and 70%).
  • The proposed algorithm outperformed traditional prediction models in terms of overall predictive performance for analyzing airline passenger behavior.

Sources:

  • "

Repurchase prediction is a vital aspect of marketing strategy and a complex decision-making task, especially in the airline industry, where data are uncertain, incomplete, and ambiguous."

-- Heilongjiang, People's Republic of China, VerticalNews, November 4, 2025.

  • Multi-classifier Evidence Ensemble Algorithm-based for Predicting Travelers Repurchases of China's Airlines. Journal of Forecasting, 2025.
  • Journal of Forecasting. Wiley-Blackwell. (www.wiley.com/; onlinelibrary.wiley.com/journal/10.1002/(ISSN)1099-131X)