Text Classification Using Label Names Only: A Language Model Self-Training Approach.
Yu MengYunyi ZhangJiaxin HuangChenyan XiongHeng JiChao ZhangJiawei HanPublished in: EMNLP (1) (2020)
Keyphrases
- language model
- text classification
- co training
- multi label
- language modeling
- n gram
- labeled data
- semi supervised learning
- probabilistic model
- text categorization
- document retrieval
- statistical language modeling
- retrieval model
- bag of words
- naive bayes
- unlabeled data
- text mining
- cost sensitive
- speech recognition
- query expansion
- machine learning
- language modelling
- feature selection
- label propagation
- information retrieval
- test collection
- training set
- translation model
- text documents
- term frequency
- statistical language models
- keywords
- context sensitive
- class labels
- cross lingual
- query terms
- image annotation
- mixture model
- text classifiers
- pseudo relevance feedback
- vector space model
- training data
- named entities
- relevance model
- co occurrence
- smoothing methods
- graph cuts
- data mining