Paper Info
Title
Exploration of Collective Pattern to Improve Location Prediction of Mobile Phone Users
Abstract
Location prediction based on cellular network traces is a very challenging task due to the randomness of the human mobility patterns. With the help of the abundant social interaction data contained in the cellular network, this paper focuson this question: How can knowing the location and the assembled and dismissed behavior of my friends be used to more accurately predict my location? In this paper, we focus on how the collective effect users' mobility. We notice an interesting rule that users tend to stay around the places where their friends are denser. Those places where friends are more crowned is chosen for a reason, either they are having a meeting there or they gathered spontaneously because of sharing office building or other resources. With known locations of friends, we established a location prediction model to make full use of those social information. In this model, we also introduced a measure of the ability of each location that weather a gathering of friends here can attract the user. The result shows that this model did improve location prediction at the average of 4.91%, and max improvement up to about 42.9%. And we summarize the feature and the difference between the kinds of users whose behavior is following his friends a lot from those user who move around more independently.
Year
DOI
Venue
2015
10.1109/SmartCity.2015.47
2015 IEEE International Conference on Smart City/SocialCom/SustainCom (SmartCity)
Keywords
Field
DocType
Location Prediction,Collective Mobility,Spontaneous Communication,Call Detail Records
Social relation,Decision tree,World Wide Web,Synchronization,Computer science,Global Positioning System,Cellular network,Notice,Mobile phone,Location prediction
Conference
Citations 
PageRank 
References 
1
0.35
11
Authors
2
Name
Order
Citations
PageRank
Chen Zhou146.94
Benxiong Huang216819.36