移动通信用户时空行为分析算法研究及系统实现  

Research and Implementation of Patio-Temporal Behavior Analysis Algorithm for Mobile Communication Users

作　　者：张振 董梅香 葛骏一 姚利民 ZHANG Zhen;DONG Meixiang;GE Junyi;YAO Liming(CETC Potevio Science&Technology Co.,Ltd.,Guangzhou Guangdong 510310,China;Beijing Normal University(Zhuhai),Zhuhai Guangdong 519087,China;China Unicom Jiangxi Branch,Nanchang Jiangxi 330096,China)

机构地区：[1]中电科普天科技股份有限公司,广东广州510310 [2]北京师范大学(珠海),广东珠海519087 [3]中国联通江西省分公司,江西南昌330096

出　　处：《通信技术》2023年第2期183-187,共5页Communications Technology

摘　　要：为了挖掘移动用户行为,提出了基于密度聚类的移动用户热点区域识别算法及融合加权频繁模式(Frequent Pattern,FP)树和最长公共子序列(Longest Common Subsequence,LCSS)算法的移动用户出行频繁轨迹模式提取方法。首先说明与思路相关的概念,其次采用密度聚类方法来识别城市的热点区域,最后融合加权FP树和LCSS算法来提取移动用户出行轨迹模式,并对算法在工程中的应用路径及应用效果进行了说明。In order to explore mobile user’s behavior,this paper proposes a mobile user hot spot recognition algorithm based on density clustering and a mobile user frequent travel trajectory pattern that combines weighted FP(Frequent Pattern)tree and LCSS(Longest Common Subsequence)algorithm.First,it explains the concepts related to ideas;then discusses density clustering methods to identify urban hot spots;finally,incorporates a weighted FP tree and LCSS algorithm to extract mobile users’travel trajectory patterns.This paper also explains how the algorithm is applied in engineering and its application effect.

关 键 词：大数据 空间密度聚类 DBSCAN算法 LCSS算法 

分 类 号：TN929.5[电子电信—通信与信息系统]