SleepNet: Attention-Enhanced Robust Sleep Prediction using Dynamic Social Networks
This work addresses sleep behavior prediction for health monitoring using mobile and wearable devices, but it is incremental as it builds on existing methods by adding social network data and attention mechanisms.
The authors tackled the problem of predicting next-day sleep duration by integrating social network contagion with physiological and phone data, using an attention mechanism to filter irrelevant connections, and demonstrated improved performance and robustness against data perturbations, with users having higher eigenvalue centrality being more vulnerable to such perturbations.
Sleep behavior significantly impacts health and acts as an indicator of physical and mental well-being. Monitoring and predicting sleep behavior with ubiquitous sensors may therefore assist in both sleep management and tracking of related health conditions. While sleep behavior depends on, and is reflected in the physiology of a person, it is also impacted by external factors such as digital media usage, social network contagion, and the surrounding weather. In this work, we propose SleepNet, a system that exploits social contagion in sleep behavior through graph networks and integrates it with physiological and phone data extracted from ubiquitous mobile and wearable devices for predicting next-day sleep labels about sleep duration. Our architecture overcomes the limitations of large-scale graphs containing connections irrelevant to sleep behavior by devising an attention mechanism. The extensive experimental evaluation highlights the improvement provided by incorporating social networks in the model. Additionally, we conduct robustness analysis to demonstrate the system's performance in real-life conditions. The outcomes affirm the stability of SleepNet against perturbations in input data. Further analyses emphasize the significance of network topology in prediction performance revealing that users with higher eigenvalue centrality are more vulnerable to data perturbations.