基于时序聚类算法的家宽大数据运营分析系统方案研究与实践

Research and Practice of Home Wide Data Operation Analysis System Based on Time Series Clustering Algorithm

随着家宽用户规模不断扩大,家宽业务泛视频化,用户对宽带网络质量要求越来越高,如何提高客户满意度,快速定位宽带故障点成为亟需解决的问题。文章从内蒙古移动家宽故障定位目前的现状痛点入手,引入时序聚类算法,提出基于时序聚类算法的家宽大数据运营分析系统,并提出该系统的两种实现方式。随后根据内蒙古家宽建设情况提出家宽大数据分析系统网络部署架构、工程实践、以及使用该技术获得的建设成效,并对未来系统建设思考提出展望。

As the scale of home broadband users continues to expand,and home broadband services become more and more video intensive,users have increasingly high requirements for broadband network quality.Improving customer satisfaction and quickly locating broadband fault points have become urgent issues to be addressed.This article starts with the current pain points of fault location in Inner Mongolia Mobiles home broadband network,proposes a home broadband big data operation analysis system based on temporal clustering algorithm,and then proposes a network deployment architecture,engineering practices,and construction results achieved using this technology for the home broadband big data analysis system based on the construction situation in Inncr Mongolia.It also proposes prospects for future system construction considerations.