On the Interpretability and Significance of Bias Metrics in Texts: a PMI-based Approach

Autor:	Valentini, Francisco, Rosati, Germán, Blasi, Damián, Slezak, Diego Fernandez, Altszyler, Edgar
Rok vydání:	2021
Předmět:	Computer Science - Computation and Language
Druh dokumentu:	Working Paper
Popis:	In recent years, word embeddings have been widely used to measure biases in texts. Even if they have proven to be effective in detecting a wide variety of biases, metrics based on word embeddings lack transparency and interpretability. We analyze an alternative PMI-based metric to quantify biases in texts. It can be expressed as a function of conditional probabilities, which provides a simple interpretation in terms of word co-occurrences. We also prove that it can be approximated by an odds ratio, which allows estimating confidence intervals and statistical significance of textual biases. This approach produces similar results to metrics based on word embeddings when capturing gender gaps of the real world embedded in large corpora. Comment: Camera Ready for ACL 2023 (main conference)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2104.06474 Zobrazit plný text záznamu View this record from Arxiv