|Weekly News Digest
November 3, 2020 — In addition to this week's NewsBreaks article and the monthly NewsLink Spotlight, Information Today, Inc. (ITI) offers Weekly News Digests that feature recent product news and company announcements. Watch for additional coverage to appear in the next print issue of Information Today. For other up-to-the-minute news, check out ITIís Twitter account: @ITINewsBreaks.
CLICK HERE to view more Weekly News Digest items.
MLconf Online Talk Explores Bias in Natural Language Processing
Lexalytics’ chief scientist, Paul Barba, is presenting “Debiasing Contextualized Word Embeddings in NLP” at MLconf Online 2020, which will be held Nov. 6. Barba will speak from 4:35 p.m. to 5:00 p.m. EST. He will cover “the efforts of Lexalytics and a team of master’s students at the University of Massachusetts Amherst Center for Data Science to reduce gender bias with contextualized word embeddings and the effects on accuracy in downstream NLP [natural language processing] tasks, challenges in working with names, and future work and applicability to other forms of latent bias.”
Lexalytics notes, “Contextualized word embeddings are state of the art in NLP, learning from large corpora of everyday language usage in order to more quickly understand how to perform new tasks. Unsurprisingly, along with general understanding of language, these embeddings pick up the biases common in society. Using the embeddings in a number of important tasks (semantic search, sentiment recognition, predictive modelling, etc.) is thus problematic, as the risk of perpetuating bias is high.”
For more information, read the press release.
Send correspondence concerning the Weekly News Digest to NewsBreaks Editor