摘要
结合大数据及机器学习的发展与实践,以安徽省O域XDR与MR数据为基础,详细阐述了数据采集过程及XDR与MR关联的实现技术,并结合实践给出了关联率与准确率影响因素分析,通过机器学习与指纹定位算法结合,利用最小欧氏距离算法不断优化指纹定位提升定位精度,实现了关联后数据的栅格化,同时基于关联数据提出了面向规划、面向网络、面向客户、面向市场的创新应用场景与方向,最后给出了建维与市场两个方向的应用案例,验证了XDR与MR关联的应用前景。
Based on the XDR and MR data records of O domain and combining with the development and practice in big data and machine learning technologies,the analysis of impact factors of correlation rate and accuracy was described in detail via practice with the help of technology of association of XDR and MR data records.By means of the integration of machine learning technology and fingerprint localization algorithm,the position accuracy was promoted constantly using the random forest algorithm,and the rasterization of associated data was realized.Simultaneously,the innovative application scenarios and directions applied to planning,network,customer and market were proposed.Finally,to verify the prospect of the association of XDR and MR data records,two examples in network construction,maintain and market were enumerated.
作者
刘韬
吴涛
王斌
LIU Tao;WU Tao;WANG Bin(Anhui Branch of China United Network Communications Co.,Ltd.,Hefei 230069,China)
出处
《电信科学》
2019年第4期120-130,共11页
Telecommunications Science
关键词
大数据
数据关联
关联成功率
关联数据应用
big data
data association
success rate of the data association
application of data association