Building a Gendered Dictionary

on YouTube ↗18m

" summary

Builds a searchable gendered-word dataset and API that grows from a few hundred Wordnik reverse-dictionary hits to about 4,000 candidates after merging a Webster JSON dump and web-scraped Collins clothing lists, then cleans them with regex (word-boundaries and optional affixes) and NLTK POS checks to drop prepositional uses (eg. “woman” as object), filters out false positives like “lioness,” and finds gender opposites using word2vec/Gensim on Google vectors with a 0.6 similarity cutoff; outcomes: a browsable index highlighting missing or imbalanced gender pairs and usable API endpoints.

AI-written summary · about

" chapters (17)

1. 0:00Intro
2. 0:23Superhero Ninja
3. 0:43Biases
4. 1:07Genderneutral pronouns
5. 2:26Building a gender addiction
6. 3:12Finding data
7. 4:13Warnock reverse dictionary
8. 5:12Static datasets
9. 5:50Regex
10. 6:46Word Boundaries
11. 8:35Finding Patterns
12. 10:32NLTK
13. 12:12Clothing Items
14. 12:54Web Scraping
15. 14:07Word Tyvek
16. 15:47Models
17. 16:36Summary

" related talks

• 34mProviding transparency about the gender visibility in newspapersDominic Herzog · DSC Europe 2022
• 38mQuestions and DiscussionWrong at the Root: Racial Bias and the Tension Between Numbers and Words in Non-Internet Data
• 35mProduct fairness testing for Machine Learning developersCherish Molezion, Qazi Mamunur Rashid · Google I/O 2022
• 20mApplying Computational Analysis to Textual Data from the Wild: a Feminist PerspectiveShauna Julia Concannon, Madeline Balaam, Emma Simpson, Rob Comber · CHI 2018
• 25mInventing Curriculum using Python and spaCyGajendra Deshpande · PyData Global 2020
• 01h 40mAfternoon Session #1: Science, Pseudo-science, Canon and BeliefJames Zou, Evelynn Hammonds, Jnoathan Kahn, Harriet Washington · Wrong at the Root: Racial Bias and the Tension Between Numbers and Words in Non-Internet Data
• 45mData Science: It's Easy as PyǃMichael Becker · PyOhio 2015
• 31mAnalyzing Rap Lyrics Using PythonJulie Lavoie · Strange Loop 2014