Recognizing activities from partially observed streams using posterior regularized conditional random fields.
Wen WenRuichu CaiZhifeng HaoXiaowei YangYang LiPublished in: Neurocomputing (2017)
Keyphrases
- conditional random fields
- partially observed
- probabilistic model
- graphical models
- information extraction
- sequence labeling
- hidden markov models
- markov random field
- random fields
- higher order
- generative model
- crf model
- named entity recognition
- maximum entropy
- image labeling
- parameter learning
- structured prediction
- posterior probability
- loopy belief propagation
- web page prediction
- approximate inference
- segmentation method
- pairwise
- gaussian process
- least squares
- structured learning
- semi markov
- graph cuts
- undirected graphical models
- active learning
- decision trees