Automatically Identifying Words That Can Serve as Labels for Few-Shot Text Classification.
Timo SchickHelmut SchmidHinrich SchützePublished in: CoRR (2020)
Keyphrases
- text classification
- n gram
- unlabeled documents
- text documents
- multi label
- distributional clustering
- training documents
- text categorization
- training corpus
- bag of words
- text mining
- feature selection
- labeled data
- text classifiers
- information theoretic
- naive bayes
- training data
- machine learning
- unlabeled data
- class labels
- semantic features
- automatic text classification
- sentiment classification
- automatically generated
- sentiment analysis
- vector space
- language modeling
- term frequency
- detection algorithm
- semi supervised learning
- visual features
- shot boundary detection
- graph cuts
- low level
- pairwise
- learning algorithm