融入位置情景的移动用户行为挖掘方法研究

Research on Mobile User Behavior Mining Integrated with Location Scenarios

移动用户为具有特殊性,即与情景的关联性.在研究融入位置情景的移动用户行为特征.首先分析了影响移动用户行为模式的情景集合,并通过调查问卷分析,提取出主要的情景因素,包括用户（user）、设备（device）、地点（location）、时间（time）和服务信息（service）.然后根据这些情景属性对移动用户行为模式进行建模,并提出了一种基于网络结构的移动用户行为模式挖掘方法.根据兴趣度的大小将移动用户行为模式划分为三个等级,即低兴趣度、中兴趣度和高兴趣度,保留兴趣度较高的行为模式,剔除兴趣度低的行为模式,然后利用数据挖掘方法挖掘出兴趣度较大移动用户行为模式的关联规则.利用这些规则对用户进行一定的推荐服务,以便为移动用户提供更优质的服务.

the behavior of mobile users is special, that is, relevance to scenarios, the purpose of this paper is to study the behavior characteristics of mobile users integrated with location scenarios. Firstly, it analyzes the scenarios that affect the behavior patterns of mobile users, and extracts the main scenarios factors through the questionnaire analysis, including the user, device, location, time and service. Then, the behavior model is modeled according to these scene attributes, and a mobile user behavior pattern mining method based on network structure is proposed. According to the degree of interestingness, the behavior patterns of mobile users are divided into three grades, namely, low interestingness, middle interestingness and high interestingness. Hold the behavior patterns with high interestingness, exclude the behavior patterns with low interestingness, and then use data mining method to mine the association rules in the behavior patterns with high interestingness. These rules can be used to recommend the service to the user, in order to provide better service for mobile users.