摘要
疫情原因使人们更加享受家庭宽带带来便捷的生活、工作服务。而家庭宽带的市场饱和,以及其对运营商存量保有的重要意义,都要求运营商提升家庭宽带精准运营能力。文章通过对山西移动原始数据进行严格的数据预处理,利用逐步回归的方法对模型变量进行筛选,利用XGBoost算法生成预测模型。与其他两种算法模型的验证集结果进行比较,本文搭建的模型具有更好的预测性能,是一个比较理想的分类模型。将潜在用户精准定位后生成高价值小区清单,促进宽带精准营销成功率提升。
The epidemic make people enjoy the convenient life and work convenient brought by family broadband.The market saturation of home broadband and its significance influence to maintain the users require operators to improve the accurate operation ability of home broadband.In this paper,through the strict data preprocessing of Shanxi Mobile's original data,the stepwise regression method is used to screen the model variables,and the XGboost algorithm is used to generate the prediction model.Compared with the results of the verification set of the other two algorithms,the model built in this paper has better prediction performance and is an ideal classification model.After the accurate positioning of potential users,the list of high-value communities have been generated to promote the success rate of family broadband.
作者
李吏豫
LI Liyu(Marketing Department of China Mobile Shanxi Co.,Ltd.,Taiyuan 030002,China)
出处
《数字通信世界》
2021年第5期94-95,共2页
Digital Communication World
