Helmholtz principle based supervised and unsupervised feature selection methods for text mining.
Melike TutkanMurat Can GanizSelim AkyokusPublished in: Inf. Process. Manag. (2016)
Keyphrases
- text mining
- helmholtz principle
- unsupervised learning
- semi supervised
- supervised learning
- unsupervised methods
- word sense induction
- weakly supervised
- supervised classification
- topic modeling
- unsupervised feature selection
- supervised methods
- discriminant projection
- supervised training
- topic models
- machine learning
- text classification
- natural language processing
- knowledge discovery
- text categorisation
- labeled data
- perceptual grouping
- information extraction
- information retrieval
- feature selection
- latent dirichlet allocation
- learning algorithm
- data analysis
- data mining
- image segmentation
- natural language
- co occurrence
- active learning
- conditional random fields
- expectation maximization