I worked on the project of Image Tagger Analysis. There we analyzed how image taggers released by the big technology companies Google, Microsoft, Amazon, Clarifai, and Imagga tag images of human beings. This study was a consecutive study based on a study published in 2018. My responsibility during the project was to extend an existing dataset by gathering image tags for portraits from the Chicago Face dataset. In addition to replicating the 2018 study results, I used NLP techniques to find biases in the data and compared the findings with the results from the previous study.
The result of the project was a publication at the ICWSM 2022 with the title Shifting Our Awareness, Taking Back Tags: Temporal Changes in Computer Vision Services’ Social Behaviors.