Selective Text Augmentation with Word Roles for Low-Resource Text Classification.
Biyang GuoSongqiao HanHailiang HuangPublished in: CoRR (2022)
Keyphrases
- text classification
- training corpus
- text data
- n gram
- text documents
- text mining
- sentence level
- document categorization
- text classifiers
- term frequency
- sentiment analysis
- text categorization
- natural language text
- text corpus
- document classification
- bag of words
- word level
- sentiment classification
- distributional clustering
- word counts
- feature selection
- text representation
- string matching
- keywords
- text segments
- machine learning
- text input
- english words
- multi label
- english text
- multiword
- linguistic information
- textual data
- lexical features
- related words
- word pairs
- word segmentation
- text retrieval
- language modeling
- printed text
- language model
- co occurrence
- syntactic categories
- chinese text
- noun phrases
- web documents
- knn
- semantic features
- concept space
- semantic information
- stop words
- natural language processing
- page layout
- printed documents
- spoken documents
- tf idf
- named entity recognizer
- punctuation marks