A Comparative Study of Using Pre-trained Language Models for Toxic Comment Classification.
Zhixue ZhaoZiqi ZhangFrank HopfgartnerPublished in: WWW (Companion Volume) (2021)
Keyphrases
- language model
- language modeling
- pre trained
- n gram
- probabilistic model
- statistical language models
- classification accuracy
- retrieval model
- query expansion
- information retrieval
- document retrieval
- test collection
- image classification
- language modelling
- supervised learning
- pattern recognition
- decision trees
- speech recognition
- feature extraction
- feature vectors
- support vector
- text classification
- feature space
- document ranking
- control signals
- multi modal
- smoothing methods
- training samples
- feature selection