Experimental tagging schema for analysis of intersectionality, bias, and reception theories.
-
Updated
Feb 21, 2024 - JavaScript
Experimental tagging schema for analysis of intersectionality, bias, and reception theories.
Code implementing differential fairness (DF) metric in: J. R. Foulds, R. Islam, K. Keya, and S. Pan. An Intersectional Definition of Fairness. 36th IEEE International Conference on Data Engineering (ICDE), 2020
Agent-based model of the effects of strong intersectionality on scientific research collaboration.
Data prep and visualizations for 'The State of US Farm Operator Livelihoods' under review at Frontiers in Sustainable Food Systems.
For a computational writing project, I created a program called "F.U.B.U. (*For Us By Us) Playlist Shuffle ✊🏿✊🏾✊🏽". My program does a lyrical analysis of the songs "Almeda" and "Don't Touch My Hair" by the artist Solange (a.k.a. Beyonce's sister).
Audrey Ford GRIDT
R Practical Using MAIHDA
Analise de Fairness em ML usando metrica ABLNI com dataset Pima Diabetes - SDK completo com visualizacoes e relatorios
Open-set discovery of intersectional bias in text-to-image models. A post-hoc extension of OpenBias (CVPR 2024): all-pairs scan of LLM-proposed attributes with bias-corrected mutual information, permutation tests, FDR control and prompt-leakage filtering.
Are you in the room? A framework for sortition and representation
An awesome list of resources at the intersection of design and research
A synthetic-data stress-test workflow and R package (imaihda) demonstrating the sensitivity of Intersectional MAIHDA summary statistics (VPC & PCV) to outcome prevalence, stratum sparsity, and detection bias.
Open educational resource on the sociology of race and racism in the United States. CC BY-NC-SA 4.0. Canonical edition on Manifold @cuny.
Unsupervised clustering applied to Curitiba's Origin-Destination survey to test whether gender and income inequalities in urban mobility emerge inductively from travel time and distance alone.
Detecting intersectional bias in LLM prompts, where demographic signal is represented by implied language rather than explicit identity terms. Fine-tuned DistilBERT router, three eval sets, and a negative result on inference-time mitigation.
Femscape is a free, open-source educational app designed to raise awareness about feminism, challenge the normalization of harm against women, explore evidence and history, and inspire meaningful action toward gender equality.
Identity Category Drift Detector (ICDD)
Detecting implied intersectional bias in LLM prompts, where demographic signal is carried by coded proxies rather than identity terms. Fine-tuned DistilBERT router, three eval sets, and a negative result on inference-time mitigation.
To associate your repository with the intersectionality topic, visit your repo's landing page and select "manage topics."