Abstract
China Mobile is the largest mobile communication operator in China, and with the promotion of its 5G applications, the issue of continuously improving its user satisfaction has become an important goal for sustainable development in the future. In this paper, data pre-processing operations were firstly performed, for data deletion and supplementation, data normalization, null filling and other steps; based on different machine learning algorithms, data feature extraction was performed to construct an effective satisfaction prediction model; entropy value method, XGBoost algorithm and lightgbm algorithm were used to train the model for prediction. The confusion matrix plot of the full variable test set was obtained by the XGBoost method, which shows that the model has some reasonableness and realistic significance.
Publisher
Darcy & Roy Press Co. Ltd.
Reference13 articles.
1. Li D.X., Wu C.L., Zou L.F. Study on the factors influencing customer satisfaction of Meituan take-out platform under O2O model Advances in Applied Mathematics 11, 5536, 2022
2. Xu JW, Yang Y. Integrated learning methods: A review of research. Journal of Yunnan University (Natural Science Edition) 40 (6), 1082-1092, 2018
3. Dong Yingying, Ge Yang, Li Kunshu, Shen Bin, Huang Shuangshuang. Research on the application of fusion model in mobile network user satisfaction prediction Post & Telecom Design Technology, 08, 2022
4. Zhang Yuchun. Research on happiness prediction and enhancement path based on stacked fusion method. Pure Mathematics 12, 1679, 202
5. Bebe Wang. Research on customer churn hierarchical prediction based on stacked integration learning in Zhejiang Mobile Company. Zhejiang University of Industry and Commerce, 2018