Automatic Gender Bias Detection in Digital Content

Abstract

The Internet and social media play an essential role in people’s daily lives. There is tremendous growth in user generated content which expedites biased views online. Gender bias is a prominent type of bias. The stereotypes of each gender today can be traced back to ancient times. Artificial Intelligence (AI) can be used to mitigate the risk of gender bias when used with caution. Our research aims to use Natural Language Processing (NLP) algorithms to identify gender biased words and their linguistic context to avoid using these biased words when referring to groups of people. Also, our solution can be used to promote unbiased business brands and product descriptions.