基于到达时间的行为模式挖掘  

Behavior Pattern Mining Based on Arrival Time

作　　者：李克华 刘志锋[1] 周从华[1] LI Kehua;LIU Zhifeng;ZHOU Conghua(School of Computer Science&Communication Engineering,Jiangsu University,Zhenjiang 212013)

机构地区：[1]江苏大学计算机科学与通信工程学院,镇江212013

出　　处：《计算机与数字工程》2020年第9期2185-2189,2219,共6页Computer & Digital Engineering

基　　金：江苏省重点研发计划(社会发展)项目(编号:BE2016630,BE2017628);无锡市卫生计生委科研项目(编号:Z201603)资助。

摘　　要：语义轨迹具有体积较小、质量较高,更好地反映移动用户的行为模式的优点。近年来不少专家开始关注语义轨迹的研究,并将其应用在了线路预测、朋友推荐、用户背景区分等领域。从语义轨迹中挖掘出代表行为习惯的频繁模式是实现这些应用的技术基础,但在一些场景中还需要分析语义点到达时间的影响,如对大学生行为模式的挖掘,在校大学生活动地点较单一,单纯考虑语义轨迹无法准确的分辨出不同的行为模式。针对这类问题,引入语义点到达时间信息,提出了一种基于到达时间的行为模式挖掘(Mining Behavior Patterns based on arrival time,MBP-at)方法。该方法的基本思路是在移动用户的语义轨迹集合中,首先挖掘每个移动用户的到达时间约束下的频繁语义行为模式,然后计算不同用户语义行为模式之间的相似度,最后通过行为模式之间的相似度对用户进行聚类,每一个聚类代表了具有相似行为模式的用户集合。实验结果表明,该方法具有有效性和高准确性。Semantic trajectory information has the advantages of small size,high quality and better reflecting the behavior patterns of mobile objects.In recent years,many experts have begun to pay attention to the research of semantic trajectory,and applied it to the application fields of line prediction,friend recommendation,user background discrimination,etc.Mining behavioral patterns from semantic trajectories is the technical basis for these applications.However,in some application scenarios,the need for the arrival time of semantic points should also be considered.To solve this problem,this paper introduces the time information of semantic points,and proposes a method of mining user semantic trajectory behavior patterns based on arrival time.The basic idea of this method is that first mining frequent semantic behavior patterns under the arrival time constraint of each mobile user,then calculating the similarity between different user’s semantic behavior patterns,and clustering the users through the similarity between behavior patterns,each clustering represents a set of users with similar behavior patterns.The experimental results show that this method is effective and accurate.

关 键 词：语义轨迹 频繁模式挖掘 到达时间 用户相似度 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]