logo Idiap Research Institute        
 [BibTeX] [Marc21]
GroupUs: Smartphone Proximity Data and Human Interaction Type Mining
Type of publication: Conference paper
Citation: Do_ISWC_2011
Publication status: Published
Booktitle: 15th annual International Symposium on Wearable Computers
Year: 2011
Month: June
Location: San Francisco, USA
Abstract: There is an increasing interest in analyzing social interaction from mobile sensor data, and smartphones 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 smartphones 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.
Projects Idiap
Authors Do, Trinh-Minh-Tri
Gatica-Perez, Daniel
Added by: [UNK]
Total mark: 0
  • Do_ISWC_2011.pdf