Multiple-instance domain adaptation for cost-effective sensor-based human activity recognition.
Aria Ghora PrabonoBernardo Nugroho YahyaSeok-Lyong LeePublished in: Future Gener. Comput. Syst. (2022)
Keyphrases
- cost effective
- domain adaptation
- multiple instance
- human activity recognition
- semi supervised learning
- activity recognition
- multiple instance learning
- semi supervised
- human activities
- labeled data
- learning problems
- unlabeled data
- low cost
- transfer learning
- real valued
- sensor data
- cross domain
- active learning
- target domain
- supervised learning
- unsupervised learning
- data sets
- test data
- machine learning
- text categorization
- training data
- data mining
- action recognition
- k nearest neighbor
- document classification
- nearest neighbor
- support vector machine
- knowledge discovery
- feature selection