Paper Info
Title
An unsupervised framework for sensing individual and cluster behavior patterns from human mobile data
Abstract
Human behavior understanding is a fundamental problem in many ubiquitous applications. It aims to automatically uncover and quantify characteristic behavior patterns in users' daily lives as well as disclose behavior clustering structure among multiple users. The key challenge is how to define a naturally interpreted representation for users' daily behavior patterns, which can be easily exploited to not only uncover the behavior similarity among multiple users but also predict users' future activities. In this paper, we define such a representation, and propose a probabilistic framework which can automatically learn it from mass amount of mobile data in unsupervised setting and exploit it to predict user activities. By an appropriate information sharing among multiple users, this framework overcomes single-user data sparsity problem and effectively identifies behavior clustering structures in a set of users. Experiments conducted on a public reality mining data set demonstrate the effectiveness and accuracy of our methods.
Year
DOI
Venue
2012
10.1145/2370216.2370241
UbiComp
Keywords
Field
DocType
human behavior understanding,multiple user,cluster behavior pattern,unsupervised framework,fundamental problem,behavior similarity,daily life,mobile data,characteristic behavior pattern,single-user data,public reality mining data,human mobile data,daily behavior pattern,graphical models
Computer science,Exploit,Human–computer interaction,Artificial intelligence,Graphical model,Cluster analysis,Reality mining,Mobile broadband,Machine learning,Information sharing,Probabilistic framework
Conference
Citations 
PageRank 
References 
33
1.58
10
Authors
2
Name
Order
Citations
PageRank
Jiangchuan Zheng11126.20
Lionel M. Ni29462802.67