The Center for Media Engagement has released a progress report on its work to find new approaches to comment moderation.
For three months, CME research associate Arielle Cardona attempted to code for incivility within online news comments in new ways –- using paid Workers in Amazon’s Mechanical Turk, undergraduate students in the Mechanical Turk sandbox, and by machine training the content analysis program WordStat.
Cardona compared how well Mechanical Turk and WordStat were at accurately detecting incivility. While she was never able to make the programs fully detect incivility, she does share her thoughts on the next possible steps in research.
Click here to read the full report.