Text classification method based on self-training and LDA topic models.
Miha PavlinekVili PodgorelecPublished in: Expert Syst. Appl. (2017)
Keyphrases
- classification method
- topic models
- text documents
- latent dirichlet allocation
- text mining
- topic modeling
- text classification
- text corpora
- probabilistic topic models
- latent semantic analysis
- k nearest neighbor
- latent topics
- co training
- knn
- support vector machine svm
- topic discovery
- probabilistic model
- support vector machine
- gibbs sampling
- variational inference
- co occurrence
- latent topic models
- lda model
- classification algorithm
- training set
- latent variables
- information retrieval
- generative model
- nearest neighbor
- knowledge discovery
- keywords
- probabilistic latent semantic analysis
- pairwise
- data sets
- neural network
- feature selection
- word sense
- semi supervised learning
- naive bayes
- named entities