基于WiFi扫描列表的用户生活模式挖掘

发布时间：2018-11-06 10:34

【摘要】：得益于装有多种传感器的智能手机的快速发展,代表用户移动过程的轨迹数据被方便的记录下来(GPS、GSM network、WiFi等)。这些蕴含用户位置信息的数据为理解用户的出行意图、移动规律和行为模式等提供了新的机会。近年来,通过用户轨迹数据来发现用户行为规律的研究成果越来越多,然而通过数据挖掘等技术手段也能非常容易的从中挖掘出各种敏感信息,这引起了人们对于隐私泄露的担忧。本文基于手机WiFi扫描列表抽象得到的用户移动轨迹,来对持机用户的移动规律进行深入研究,目标是在保护用户隐私的基础上挖掘用户的生活模式。本文的主要贡献有：1)提出了一种使用图模型来表示用户移动性的方法,通过对WiFi扫描列表聚类得到用户访问地点,然后基于用户在地点间的轨迹来构建移动图。2)在用户移动图上,一方面基于用户频繁轨迹来发现用户群体的移动模式,另一方面通过用户在活动区域上的行为特征来进一步刻画用户移动性。3)设计了一种基于地点时间特征的聚类质心叠加方法来识别所有用户的家和工作地点,通过分析用户在家和工作地的行为来理解用户作息规律。
[Abstract]:Thanks to the rapid development of smart phones with multiple sensors, track data representing the user's movement process are conveniently recorded (GPS,GSM network,WiFi et al.). These data, which contain user location information, provide a new opportunity to understand the user's travel intentions, movement patterns and behavior patterns. In recent years, there are more and more research results of discovering user behavior rules through user trajectory data. However, it is very easy to mine all kinds of sensitive information through data mining and other technical means. This raises concerns about privacy leaks. Based on the mobile path abstracted from the mobile phone WiFi scanning list, this paper makes an in-depth study on the mobile law of the mobile user. The goal is to mine the user's life pattern on the basis of protecting the user's privacy. The main contributions of this paper are as follows: 1) A graph model is proposed to represent the mobility of the user. The location of the user is obtained by clustering the WiFi scan list. Then the mobile map is constructed based on the user's local-to-place locus. 2) on the user's mobile map, on the one hand, based on the user's frequent locus, the mobile mode of the user group is discovered. On the other hand, a clustering centroid superposition method based on location time feature is designed to identify the home and work place of all users. By analyzing the behavior of the user at home and at work, we can understand the law of the user's working and rest.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.13

【相似文献】