Reading Between the Demographic Lines: Resolving Sources of Bias in Toxicity Classifiers.
Elizabeth ReichertHelen QiuJasmine BayrootiPublished in: CoRR (2020)
Keyphrases
- training data
- training set
- linear classifiers
- supervised classification
- test set
- bias variance decomposition
- decision trees
- support vector
- classifier ensemble
- feature selection
- straight line
- naive bayes
- hough transform
- information sources
- multiple sources
- feature set
- training samples
- multiple classifiers
- individual classifiers
- ensemble classifier
- classification systems
- data sources
- classification models
- line segments
- ensemble learning
- feature space
- machine learning methods
- trained classifiers
- machine learning