Designing an AI to grade for curiosity isn’t easy; psychologists have been trying to boil curiosity down to a definable metric for generations. At Packback, our research team has been working for the past year to develop an algorithm that measures quality of the posts on the Packback platform. While we’re not quite ready to announce that the Packback Curiosity Score is the definitive metric to objectively measure curiosity (that would be BIG news!)…we are very excited about the accuracy with which it can score effort, presentation, and credibility of every question and response on Packback!
The Packback team loves our feature, Real-Time Coaching, to shorten the feedback loop and deliver coaching when it matters most; while a student is still writing their post.
In Packback Questions communities across the nation, thousands of questions are posted per week. That is a lot of questions flowing through the platform, which could make it easy to miss some of the best posts. The new Featured Tab places the best discussions in every community in their own easy-to-navigate tab, so they are always only one click away.
In the past year, Packback’s team of Community Managers have moderated over 350,000 questions and responses and maintained the health of all of the Curiosity Communities on Packback Questions.
How did they do it?
Through careful manual moderating and coaching around the Packback Community Guidelines, supported by Crowdsourced Flagging (where students and professors help “flag” offending posts in their own communities) and augmented by Packback’s proprietary Community Health Algorithm (that notifies our Community Managers of all posts that may not meet the Community Guidelines).