基于XGBoost的5G终端换机品牌预测  

Brand Prediction for 5G Mobile Replacement Based on XGBoost

作　　者：吴瑜 严李强 张福豪 方晓捷 WU Yu;YAN Liqiang;ZHANG Fuhao;FANG Xiaojie(School of Information Science and Technology,Tibet University,Lhasa 850000,China)

机构地区：[1]西藏大学信息科学技术学院,拉萨850011

出　　处：《科技和产业》2024年第18期231-239,共9页Science Technology and Industry

基　　金：国家自然科学基金(62161047);西藏自治区科技厅科技重大专项(XZ202201ZD0006G02)。

摘　　要：根据中国信通院发布的《2023年12月国内手机市场运行分析报告》显示,2023年1—12月国内市场5G手机出货量占同期手机出货量的82.8%,同比增长11.9%。随着5G技术的全球普及和5G手机终端的出货量逐步提高,各大企业的竞争日趋激烈;如何准确预测用户的5G终端换机品牌偏好已成为企业当前的研究热点。为解决预测最适合用户的5G终端换机品牌问题,通过对用户的个人数据和历史行为数据进行预处理、特征提取、特征衍生,选取重要性最高的特征,使用XGBoost模型和贝叶斯优化完成5G终端换机品牌预测模型的构建。结果表明,模型在用户的5G终端换机品牌预测方面具有较高的准确率,企业可以依此制定针对性的营销策略,提升用户黏性,从而提升产品的市场竞争力。According to the“December 2023 Domestic Mobile Phone Market Operation Analysis Report”released by the China Academy of Information and Communications Technology,shipments of 5G mobile phones in the domestic market reached 82.8%of the total mobile phone shipments during the same period in 2023,marking an increase of 11.9%.As 5G technology continues to proliferate globally and shipments of 5G mobile terminals rise steadily,the competition among major enterprises has notably intensified.Accurately predicting consumer preferences for 5G terminal brands during upgrades has become a critical area of focus for business research.Accounting to these issues,users’personal data and historical behavior to select key features were thoroughly preprocessed and analyzed.XGBoost model combined with Bayesian optimization to was used to develop a predictive model that could accurately determine users’preferred brands for their 5G terminal replacements.The results show that this model achieves high accuracy in forecasting consumer brand preferences in the 5G terminal replacement market.This capability allows enterprises to craft precise marketing strategies aimed at enhancing user retention,thereby significantly boosting their competitive edge in the market.

关 键 词：用户行为预测 手机品牌 XGBoost模型 机器学习 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]