Necessity and Sufficiency for Explaining Text Classifiers: A Case Study in Hate Speech Detection

TL;DR

This paper introduces a new feature attribution method for text classifiers that uses necessity and sufficiency scores to provide more informative explanations, especially in hate speech detection, revealing biases and error sources.

Contribution

It proposes a transparent, perturbation-based approach to compute necessity and sufficiency scores, enhancing interpretability of text classifier explanations.

Findings

Different necessity and sufficiency scores reveal classifier biases.

The method exposes sources of false positives in hate speech detection.

Explicit perturbations improve explanation transparency.

Abstract

We present a novel feature attribution method for explaining text classifiers, and analyze it in the context of hate speech detection. Although feature attribution models usually provide a single importance score for each token, we instead provide two complementary and theoretically-grounded scores -- necessity and sufficiency -- resulting in more informative explanations. We propose a transparent method that calculates these values by generating explicit perturbations of the input text, allowing the importance scores themselves to be explainable. We employ our method to explain the predictions of different hate speech detection models on the same set of curated examples from a test suite, and show that different values of necessity and sufficiency for identity terms correspond to different kinds of false positive errors, exposing sources of classifier bias against marginalized groups.