Word embeddings serve as novel objects for sociological analysis of bias
Categorization
- corroborated: 3 independent passages support this claim
- XREF: connects to: Word embeddings’ gender skew correlates with real occupational gender ratios; Large-scale English text corpora reflect real-world societal biases; Even massive, comprehensive text datasets reflect biases present in the real world (+5 more)
Related
- Large-scale English text corpora reflect real-world societal biases (similarity 0.95)
- Hundreds of papers missed blatant sexism in word embeddings (similarity 0.89)
- Common language carries more gender bias than objective metrics show (similarity 0.87)
- Word embeddings’ gender skew correlates with real occupational gender ratios (similarity 0.82)
- Machine learning increasingly intertwines technical problems with human and social questions (similarity 0.80)
Library Connections
- CLAUDE
#complicates(0.88) — The second channel — because embeddings miss the best links
Nearest-neighbor retrieval systematically finds the *obviou…
- Even massive, comprehensive text datasets reflect biases present in the real world
#corroborates(0.82) - Large-scale English text corpora reflect real-world societal biases
#corroborates(0.80)
Broader Connections
- Machine learning increasingly intertwines technical problems with human and social questions
#expands(0.78) - Structural Islamophobia and structural racism share embedded institutional stereotypes
#exemplifies(0.60) — Khaled A. Beydoun, American Islamophobia_ Understanding…, 130