MA
Maria Antoniak
@mariaa.bsky.social
Pioneer Centre for AI @ University of Copenhagen, incoming Asst Prof @ CU Boulder CS, #NLProc and #CulturalAnalytics, interested in narratives, healthcare, online communities. Previously Ai2, Cornell, Twitter, Microsoft. maria-antoniak.github.io
1k followers579 following493 posts
Reading a new paper and people are still using Empath. I really do not recommend Empath as a tool for measuring lexical categories and biases: aclanthology.org/2021.acl-lon...
Bad Seeds: Evaluating Lexical Methods for Bias Measurement
Maria Antoniak, David Mimno. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume ...
Or to put it differently: if you're going to use a lexicon, explain why that lexicon is a good choice for your data and give some indication that you know what kinds of words are included in the lexicon.
MA
Maria Antoniak
@mariaa.bsky.social
Pioneer Centre for AI @ University of Copenhagen, incoming Asst Prof @ CU Boulder CS, #NLProc and #CulturalAnalytics, interested in narratives, healthcare, online communities. Previously Ai2, Cornell, Twitter, Microsoft. maria-antoniak.github.io
1k followers579 following493 posts