Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification.
Daniel BorkanLucas DixonJeffrey SorensenNithum ThainLucy VassermanPublished in: CoRR (2019)
Keyphrases
- text classification
- feature selection
- bag of words
- synthetic data
- machine learning
- text data
- text categorization
- text mining
- naive bayes
- n gram
- labeled data
- text classifiers
- semantic features
- sentiment analysis
- data cleaning
- document classification
- language modeling
- knn
- database
- evaluation metrics
- text documents
- term frequency
- multi label
- similarity measure
- feature space
- similarity metrics