Self-supervised learning for human activity recognition using 700,000 person-days of wearable data.
Hang YuanShing ChanAndrew P. CreaghCatherine TongAidan AcquahDavid A. CliftonAiden R. DohertyPublished in: npj Digit. Medicine (2024)
Keyphrases
- human activities
- data sets
- sensor data
- activity recognition
- reinforcement learning
- data analysis
- human computer interaction
- prior knowledge
- data sources
- data points
- unsupervised learning
- ubiquitous computing
- knowledge discovery
- learning process
- sensor networks
- daily activities
- human knowledge
- wearable sensors
- shape model
- text classification
- training data
- image sequences
- learning algorithm