Language selection

Government of Canada / Gouvernement du Canada

Search

Towards procedural fairness: uncovering biases in how a toxic language classifier uses sentiment information

From National Research Council Canada

Download
  1. (PDF, 676 KiB)
AuthorSearch for: 1; Search for: 1; Search for: 1; Search for: 1
Affiliation
  1. National Research Council of Canada. Digital Technologies
FormatText, Article
ConferenceFifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, December 7, 2022, Abu Dhabi, United Arab Emirates (Hybrid)
Abstract
Previous works on the fairness of toxic language classifiers compare the output of models with different identity terms as input features but do not consider the impact of other important concepts present in the context. Here, besides identity terms, we take into account high-level latent features learned by the classifier and investigate the interaction between these features and identity terms. For a multi-class toxic language classifier, we leverage a concept-based explanation framework to calculate the sensitivity of the model to the concept of sentiment, which has been used before as a salient feature for toxic language detection. Our results show that although for some classes, the classifier has learned the sentiment information as expected, this information is outweighed by the influence of identity terms as input features. This work is a step towards evaluating procedural fairness, where unfair processes lead to unfair outcomes. The produced knowledge can guide debiasing techniques to ensure that important concepts besides identity terms are well-represented in training datasets.
Publication date
PublisherAssociation for Computational Linguistics
Licence
In
LanguageEnglish
Peer reviewedYes
Export citationExport as RIS
Report a correctionReport a correction (opens in a new tab)
Record identifier0d5adc31-9270-4e9d-ab5e-6e9f570708dc
Record created2023-02-03
Record modified2023-02-14
Date modified: