A Dining Context-Aware System with Mobile and Wearable Devices

With development of various sensors attached to mobile and wearable devices, recognizing user´s current context and giving an appropriate service come to hot issue. In this paper, we propose the context-aware system recognizing user´s dining context that can occur within a great variety of contexts. The model uses low-level sensor data from mobile and wristwearable devices that can be widely available in daily life. To cope with innate complexity and fuzziness in high-level contexts like dining, a context model represents the related contexts systemically based on 4 components of activity theory and 5 W´s, and tree-structured Bayesian network can recognizes the dining context probabilistically. To verify the proposed system, we collected 383 minutes of data from 4 people in a week and found that the proposed method outperforms the conventional machine learning methods, as to accuracy (94.57%). Also we built an Android application to investigate its practicality, and conducted a scenario-based test to investigate the effect of individual context for recognition.