Disinformation: Starter Kit - Bias

Bias is prevalent in every aspect of our lives. Our brains are hardwired to categorize things we encounter in order to make sense of the complicated world around us. However, biases can cause us to form prejudices against others, which allows for egregious inequalities to form between different demographics.

While bias comes in many forms, bias words in writing is one form. Implicit bias in letter writing or evaluations negatively affects individuals at every stage of their career.

In this challenge, we are inviting to build a solution for detecting bias in writings such as letter of recommendations, Job Descriptions etc with respect to gender and race for promoting equity. You can choose any topic of your choice. The sample solution kit helps to detect gender bias.

Instruction to Run

Follow below instructions to run the solution.

1. Locate and open the gender-bias.ipynb notebook from the Jupyter Notebook browser window.
2. Execute cells in the notebook by selecting Cell --> Run All from Menu bar

For running it with your text,

1. Open letterofRecW file from the location data/input from gender-bias.ipynb location.
2. Update text in the letterofRecW file.
3. Execute cells in the notebook by selecting Cell --> Run All from Menu bar.
4. Output will be stored in a file gender-biased-words.txt in the location data/output. Output text is in json format. Output data format is: name - detector name. e.g. "Terms biased towards women" summary - summary of the detected bias flags - flag the detected bias words. e.g. "leader" You can additionally create your own detectors for race and dictionary dataset as well as other enhancements for additional score.

Troubleshooting

1. While running batch file, if you encounter Windows protection alert, select More info --> Run anyway
2. During kit installer, if you encounter Windows security alert, click Allow

For a detailed tutorial on installing & executing the solution as well as learning resources including training & certification opportunities, please visit the OpenWeaver Community

Development Environment

VSCode and Jupyter Notebook are used for development and debugging. Jupyter Notebook is a web based interactive environment often used for experiments, whereas VSCode is used to get a typical experience of IDE for developers. Jupyter Notebook is used for our development.

Text Mining

Libraries in this group are used for analysis and processing of unstructured natural language. The data, as in its original form aren't used as it has to go through processing pipeline to become suitable for applying machine learning techniques and algorithms.

Testing

The libraries listed here can be used for unit testing as well as integration testing

Support

If you need help using this kit, you may reach us at the OpenWeaver Community.