Technologies for Information Systems (TIS) Final Project.
-
Updated
Mar 14, 2023 - Jupyter Notebook
Technologies for Information Systems (TIS) Final Project.
Implementation of alternation functions in C++ for detection and evaluation of bias in ML
Pytorch implementation of 'Explaining text classifiers with counterfactual representations'
Replicate, update implementation of sent-bias: Social Bias in Sentence Encoders
Report for INFO4900 Independent Research under Prof. Dawn Schrader. Surveyed bias detection and mitigation methods in language models. Identified emerging Language Model tasks where existing mechanisms fail. Designed a novel fairness test and proposed a framework to update large language models when what society considers fair changes.
Free tools to help enterprises and individual audit biases in their algorithms. Support legislation such as the NYC AEDT Law.
PMD ruleset for detecting racially charged or biased language
A challenge set comparing male/female diacritization in Hebrew
A bias detection system that takes text as input and analyses Sentence-wise. Developed by Anand Chauhan and Vasu Jain as part of our bachelor's degree
Demographically-Informed Prediction Discrepancy Index: Early Warnings of Demographic Biases for Unlabeled Populations
Estimating the age from images while tacking the bias with respect to the protected attributes (Age, Gender, Ethnicity, Face Expression)
This project aims to detect the presence of bias in news articles
A demonstration of detecting and mitigating bias in AI.
GitHub repository investigating gender bias in employee attrition prediction. Mitigate bias, promote fairness in workforce analytics. Code and resources for addressing gender disparities, improving decision-making
Aimed to predict anxiety among Canadian Population during Covid 19
Reproducing Pro-Publica's story Auditing for Bias against African American Individuals
Repos for On the Reliability and Explainability of Language Models for Program Generation
Bias Buster is a Chromium extension that enables users to comments and identify bias and prejudice on arbitrary webpages.
🛒 Webscraper used to detect bias in Amazon product reviews for a product. Bias is determined through an original and rigorous algorithm, with the goal of producing a new, corrected, product star review from 1 to 5 based only on reviews that are considered to be 'unbiased'.
Add a description, image, and links to the bias-detection topic page so that developers can more easily learn about it.
To associate your repository with the bias-detection topic, visit your repo's landing page and select "manage topics."