Analysis Of Student's Behavior Trajectory Based On Mobile Client Location Information

People's daily production and life will produce a lot of footprints,The location information appears to be disorganized,but in fact there are a lot of regular patterns can be found,It contains a wealth of information about the spatial structure and behavior rules.Through mining it's hidden knowledge,it can serve the smart city.Especially in the study of College Students,the behavior itself of students is in high regularity.So we can mining their regular patterns in the use of trajectory data,then the consequences can help school administrators to make strategic decisions for the efficient and intelligent management on students.The traditional way to gain trajectory data is mainly the investigation of activity diary,but it's precision accuracy is very low,and meanwhile it brings burden for interviewees.With the development of mobile positioning technology and the popularity of smart phones,there derives many new technical means to gain trajectory data of users,one of them is to collection smartphone positioning data.In order to analyze the daily behavior pattern of college students,the following research work has been done on data acquisition,trajectory management,stay point identification and behavior trajectory analysis in this paper.(1)The data acquisition software based on Android platform is designed and developed.And the project requirements,design philosophy,design process,UI design,functional design and database design are described in detail.Then Chang'an University graduate students are recruited as volunteers to install this software to acquire data.(2)The WebGIS visualization analysis platform is designed and developed based on time geography framework.It is secondary development of Baidu Map for management,query,analysis and visualization of trajectory data.(3)The method of coordinate transformation for multitudinous location data is programmed,and the method of calculating distance between the two points of the sphere is programmed.The stay point extraction algorithm is designed and implemented on the basis of the existing DBSCAN clustering algorithm to confirm the time and location of the stay point.Then the trajectory data simulation system is designed and developed to verify the algorithm.Finally,the Baidu Geocoder service class is used to obtain POI information of stay point.(4)The behavior pattern of college students is defined at first.The trajectory data of onemonth from one of the volunteers is chosen as research object to analyze the student's behavior trajectory pattern.First,the behavior trajectory of a certain day is analyzed.Then,this paper makes a comparative analysis between frequent behavior patterns and frequent sequential behavior patterns.Finally,the law of student's behavior could be unscramble by charts.