How well do hate speech, toxicity, abusive and offensive language classification models generalize across datasets?
Paula FortunaJuan Soler CompanyLeo WannerPublished in: Inf. Process. Manag. (2021)
Keyphrases
- classification models
- training data
- feature selection
- language acquisition
- text to speech synthesis
- decision trees
- text to speech
- spoken language
- natural language
- models built
- imbalanced data
- speech recognition
- english text
- decision tree algorithm
- software quality classification
- attribute selection
- database
- feature set
- benchmark datasets
- feature subset
- learned models
- optimization method
- image classification
- artificial neural networks
- training set
- data sets