stigsfoot / examingBiasInAI

This is a short project to highlight how easy it is to create an ML model that exemplifies bias. I also make an attempt to mitigate said bias with a disparate impact removal algorithm provided by the AI Fairness 360 toolkit.

Examining Bias in AI

Predicting loan approval and examining the bias in doing so

This is a short project to highlight how easy it is to create an Machine Learning model that amplifies bias. I also make an attempt to mitigate said bias with a disparate impact removal algorithm provided by the AI Fairness 360 toolkit.

I have written a short Medium article associated with the project, in which I explain the goals of this exercise as well as explore the meaning of its results.

Read the article here.

Thanks!