Multimodal autoencoder: A deep learning approach to filling in missing sensor data and enabling better mood prediction.
Natasha JaquesSara TaylorAkane SanoRosalind W. PicardPublished in: ACII (2017)
Keyphrases
- sensor data
- deep learning
- restricted boltzmann machine
- sensor networks
- unsupervised learning
- data streams
- unsupervised feature learning
- machine learning
- health monitoring
- mental models
- raw sensor data
- sensor measurements
- multiple sensors
- sensor readings
- weakly supervised
- decision support system
- neural network
- expert systems
- computer vision
- data sets
- principal component analysis