Sourcing Comments

My Week In Three Words: Sourcing Comments and IRB

What I’ve been working on: This week I’ve been working on sourcing comments that were labeled by toxicity and several other metrics from this dataset. A huge thank you to Dr. Kumar for giving us access to the dataset and supporting us on building off their team’s work! I wrote a script to convert the JSON file to a CVS file in Python, then wrote another script to find comments that fit our specific criteria. Basically we need to get a certain number of comments from each category (for example: insult, threat, sexual harassment, etc.) and then also comments falling along the Perspective API score and individual toxicity ratings to use as comments in a simulated Twitter feed and as examples for our slider configurations in an exercise for participants during our study. Additionally, I wrote up a methods/study description which described how we were planning to conduct our study in detail. I also finished filling in some details for our IRB form–the first one I’ve officially completed! Finally toward the end of the week I began playing around with different types of slider configuration mock-ups.

What I’ve learned:

1. How to convert from JSON to CSV
2. I went to a seminar where Benjamin Xie and Kentrell Owens spoke about their work. It was really interesting because Kentrell did presented on one of his studies about the surveillance of communication with incarcerated people, and it was the first time I was exposed to this sector of HCI. Benjamin talked about his work on how equity can be improved in code.org’s CS curriculum. I am so excited to look more into Benjamin and Kentrell’s work and would definitely encourage others to do so!

What I’m curious about:

1. I’m curious about learning more about the intersection of criminal justice and computation in HCI
2. I’m also curious about collapse informatics. I was fortunate to be able to meet with Benjamin and he actually introduced me to this topic and sent me this paper