Using Word Sense as a Latent Variable in LDA Can Improve Topic Modeling.
Yunqing XiaGuoyu TangHuan ZhaoErik CambriaThomas Fang ZhengPublished in: ICAART (1) (2014)
Keyphrases
- topic modeling
- topic models
- latent variables
- latent dirichlet allocation
- probabilistic topic models
- word sense
- probabilistic latent semantic analysis
- latent topics
- text mining
- latent variable models
- probabilistic model
- co occurrence
- text documents
- text classification
- collaborative filtering
- generative model
- prior knowledge
- data mining
- word sense disambiguation
- random variables
- latent semantic analysis
- active learning
- feature extraction
- learning algorithm