Research on User Behavior Prediction Based on XGBoost Model Enhanced by Energy Valley Optimizer
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Overfitting and data imbalance pose challenges to traditional machine learning models. In order to improve the prediction accuracy, this study proposes the EVO-XGBoost algorithm, which innovatively integrates the Energy Valley Optimizer (EVO) and the XGBoost model and is interpretable. Before running the algorithm, the Synthetic Minority Over-sampling Technique (SMOTE) is first used to address the class imbalance problem, and then the EVO is applied to optimize the hyperparameters of XGBoost to enhance the model’s ability to learn and recognize complex data. The experimental results show that the F1-Score of EVO-XGBoost in the test reaches 0.662, which is 0.014 higher than that of the unoptimized XGBoost model, and significantly improves the accuracy and reliability of the prediction. Finally, SHAP analysis identifies the key features influencing purchase decisions. The research findings provide an effective method for tourism e-commerce companies to accurately predict user behaviors and determine marketing strategies.
Fields of Study & Subjects
Imbalanced Data Classification Techniques
Customer churn and segmentation
Data Mining and Machine Learning Applications
Publication details
- Journal / source
- IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
- Pages
- 237–244
- Publication date
- 28 January 2026
- Language
- EN