GroupUs: Smartphone Proximity Data and Human Interaction Type Mining
- 1 June 2011
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
There is an increasing interest in analyzing social interaction from mobile sensor data, and smart phones are rapidly becoming the most attractive sensing option. We propose a new probabilistic relational model to analyze long-term dynamic social networks created by physical proximity of people. Our model can infer different interaction types from the network, revealing the participants of a given group interaction, and discovering a variety of social contexts. Our analysis is conducted on Bluetooth data sensed with smart phones for over one year on the life of 40 individuals related by professional or personal links. We objectively validate our model by studying its predictive performance, showing a significant advantage over a recently proposed model.Keywords
This publication has 17 references indexed in Scilit:
- GDC: Group Discovery Using Co-location TracesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2010
- Modeling relational events via latent classesPublished by Association for Computing Machinery (ACM) ,2010
- Inferring friendship network structure by using mobile phone dataProceedings of the National Academy of Sciences of the United States of America, 2009
- Sensible Organizations: Technology and Methodology for Automatically Measuring Organizational BehaviorIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2008
- Discovery of activity patterns using topic modelsPublished by Association for Computing Machinery (ACM) ,2008
- The dynamics of a mobile phone networkPhysica A: Statistical Mechanics and its Applications, 2008
- Extracting Places and Activities from GPS Traces Using Hierarchical Conditional Random FieldsThe International Journal of Robotics Research, 2007
- Reality mining: sensing complex social systemsPersonal and Ubiquitous Computing, 2005
- Finding scientific topicsProceedings of the National Academy of Sciences of the United States of America, 2004
- Estimation and Prediction for Stochastic BlockstructuresJournal of the American Statistical Association, 2001