Algorithmic Bias: When stigmatization becomes a perception: The stigmatized become endangered

In this study, the author examines how perceived stigmatization endangered the stigmatized groups within a society or community. Thus, he goes back in history to dig deep into the sources of perceived stigmatization associated with the black race and how perceived stigmatization has emigrated into AI tools and machine outputs - subjecting vulnerable communities to hypervisibility by exposing them to systems of racial surveillance. To justify the study goal, he conducted a summarized text analysis on racial stigmatization using Twitter hashtags ∈ { black people, blackness, Africa, African-Americans}, all coined out of the Twitter Users’ perception of the subject and hypothesized to find high negative sentiment correlation of stigmatization perspective in association with black race and Africa. He finds that Black people are associated with Africa and have a strong negative sentiment correlation with - poorness, crime, death, abuses (stupid), among others, and a subject of racist scum and racism. Similarly, there is a weak negative sentiment correlation with being - bad, abused (such bitch), hate, violence, and protest. He also finds similar strong and weak negative sentiment correlations with other hashtags. He discusses the danger of racial stigmatization and proposes a cycle of ethical algorithmic development & deployment and recommendations.