Research on Urban E-commerce Logistics Volume Based on XGBoost Regression Tree
DOI:
https://doi.org/10.53469/wjimt.2023.06(05).07Keywords:
Urban logistics, Logistics volume prediction, XGBoostAbstract
With the popularization of the Internet, online shopping has become prevalent, leading to rapid development in the logistics industry. In order to predict the volume of logistics between cities, this paper proposes a city logistics volume prediction model based on the XGBoost algorithm and validates it using real city logistics datasets. The experiment demonstrates that this prediction model can accurately forecast the logistics volume between different cities, exhibiting good predictive performance and robustness. Research on logistics volume prediction is beneficial for optimizing urban logistics planning and enhancing logistics efficiency.
References
Yang Yue: Research on Yunda Express Business VolumeForecast Based on RBF Neural Neywork (Master of Engineering, Harbin University of Science and Technology, China 2022).
Xu Rongbin, Wang yeguo, Wang futian, et al. Prediction of package volume based on improved POS-BP [J]. Computer Integrated Manufacturing Systems, 2018, 24 (7): 9.
Chen Tianqi, C. Guestrin. ""XGBoost: A Scalable Tree Boosting System."" ACM (2016).
Friedman, Jerome H. Greedy Function Approximation: A Gradient Boosting Machine [J]. Annals of Statistics, 2001, 29 (5): 1189-1232.
Ron, K. A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, 1995 American Association for Artificial Intelligence, 1995.