SRTM: a supervised relation topic model for multi-classification on large-scale document network.
Chunshan LiHua ZhangDianhui ChuXiaofei XuPublished in: Neural Comput. Appl. (2020)
Keyphrases
- topic models
- text documents
- latent topics
- document classification
- topic discovery
- weakly supervised
- latent dirichlet allocation
- topic modeling
- unsupervised learning
- supervised learning
- feature selection
- text mining
- machine learning
- pattern classification
- text classification
- supervised topic models
- support vector machine
- network structure
- latent variables
- image classification
- statistical topic models
- information retrieval
- relevance model
- co occurrence
- lda model
- keywords
- text classifiers
- support vector
- generative model
- feature space
- training set
- active learning
- class labels
- probabilistic model
- data mining
- generative process
- model selection