Can We Ditch Feature Engineering? End-to-End Deep Learning for Affect Recognition from Physiological Sensor Data.
Maciej DziezycMartin GjoreskiPrzemyslaw KazienkoStanislaw SaganowskiMatjaz GamsPublished in: Sensors (2020)
Keyphrases
- end to end
- deep learning
- sensor data
- feature engineering
- affect recognition
- machine learning
- emotional state
- affective computing
- dependency parsing
- physiological signals
- behavioral data
- text classification
- health monitoring
- sensor networks
- natural language processing
- unsupervised learning
- data streams
- intelligent tutoring systems
- labeled data
- affective states
- weakly supervised
- human computer interaction
- mental models
- human activities
- facial expressions
- artificial intelligence
- information extraction
- knowledge representation
- text mining
- decision support system
- active learning
- heart rate
- pairwise
- pattern recognition
- multiscale
- feature extraction
- feature selection
- learning algorithm