Global Data Science Forum

Expand all | Collapse all

AI Code Pattern: Identify and remove bias from AI Models

  • 1.  AI Code Pattern: Identify and remove bias from AI Models

    Posted Tue March 02, 2021 04:32 PM

    "Fairness in data and machine learning algorithms is critical to building safe and responsible AI systems. While accuracy is one metric for evaluating the accuracy of a machine learning model, fairness gives you a way to understand the practical implications of deploying the model in a real-world situation.

    In this code pattern, you use a diabetes data set to predict whether a person is prone to have diabetes. You'll use IBM Watson® Studio, IBM Cloud Object Storage, and the AI Fairness 360 Toolkit to create the data, apply the bias mitigation algorithm, then analyze the results."

    Original code pattern



    ------------------------------
    William Roberts
    ------------------------------


  • 2.  RE: AI Code Pattern: Identify and remove bias from AI Models

    Posted Wed March 03, 2021 02:11 PM
    I've tried it and it's awesome 👏 


    ------------------------------
    Safiyya Kareem
    ------------------------------



  • 3.  RE: AI Code Pattern: Identify and remove bias from AI Models

    Posted Wed March 03, 2021 07:05 PM
    Fantastic! Have you tried using AIF 360 on any other datasets as well?

    ------------------------------
    William Roberts
    ------------------------------


  • 4.  RE: AI Code Pattern: Identify and remove bias from AI Models

    Posted Thu March 04, 2021 03:11 AM
    I have tried it with Kaggle datasets :D

    ------------------------------
    Safiyya Kareem
    ------------------------------