摘 要:针对公众住宅小区千兆宽带用户规模不确定导致预覆盖 OLT 设备投资建设不精确的现状,提出了一种基于随机森林算法及地理围栏进行小区千兆用户规模预测的方案。方案使用小区场景围栏完成地理数据、房地产中介数据与运营商数据的匹配,通过小区内楼宇栋数、住户数、房屋出租均价、运营商宽带用户数等指标对小区千兆用户规模进行预测。对比实验证明,使用随机森林算法进行千兆用户规模预测具有较高的准确性。
关键词:随机森林;地理围栏;千兆宽带;用户预测
DOI:10.19850/j.cnki.2096-4706.2023.01.016
中图分类号:TP18 文献标识码:A 文章编号:2096-4706(2023)01-0061-03
Prediction of Gigabit Broadband Subscriber Scale Based on Random Forest and Geo-fencing
CHE Weisong1, QI Jing2, BAI Wenqi 2
(1.Beijing Telecom Planning & Designing Institute Co., Ltd., Beijing 100048, China; 2.China United Network Communications Group Co., Ltd., Beijing 100033, China)
Abstract: In light of the current situation of inaccurate investment and construction in pre-coverage OLT equipment due to the uncertainty of the scale of gigabit broadband subscriber in public residential quarters, this paper proposes a scheme for gigabit subscriber scale prediction based on Random Forest algorithm and Geo-fencing. The scheme uses the Geo-fencing of the residential area to complete the matching among geographical data, real estate agency data and operator data, and predicts the scale of gigabit subscriber through indicators such as the number of buildings in the residential area, the number of households, the average house rental price, and the number of operators' broadband subscribers. The comparative experiments prove that the Random Forest algorithm has higher accuracy in predicting the scale of gigabit subscribers.
Keywords: Random Forest; Geo-fencing; gigabit broadband; subscriber prediction
参考文献:
[1] 中华人民共和国工业和信息化部 .2022 年 1-5 月份通信业经济运行情况[EB/OL].https://www.miit.gov.cn/gxsj/tjfx/txy/art/2022/art_84d43f3ff32c457dabf66dcedfa7996f.html.
[2] 鞠明山 . 地理围栏技术发展现状及网络信息安全问题探究[J]. 江西通信科技,2018(3):39-41.
[3] 吕红燕,冯倩 . 随机森林算法研究综述 [J]. 河北省科学院学报,2019,36(3):37-41.
[4]] CHAWLA N V,BOWYER K W,HALL L O,et al. SMOTE:Synthetic Minority Over-Sampling Technique [J].Journal of Artificial Intelligence Research,2002,16:321-357.
[5] 乔健,诸佳慧,严康桓 . 基于随机森林 CART 特征选择改进算法的电信客户流失预测模型 [J]. 电信工程技术与标准化,2022,35(3):78-82.
作者简介:车维崧(1990—),男,汉族,北京人,助理工程师,硕士,研究方向:宽带政企接入网络规划;祁静(1988—),女,汉族,北京人,工程师,硕士,研究方向:宽带政企接入网数字化运营;白文其(1977—),男,满族,北京人,高级工程师,本科,研究方向:传输及宽带网络技术。