Synopsis
This text extends on our analysis into predicting engagement in online platforms, using deep learning. We discovered predicting person behaviour was depending on having adequate historic information, which was in line with present analysis into the realm.
Non-paid platforms usually search to encourage and reward participation via badges, medals and on-line incentives. Though they are often efficient, they usually yield unintended behaviours and blended outcomes. Each Coursera and StackOverflow for instance have witnessed “steering”. Customers usually work to attain a badge then disengage from the platform. Though this boosts engagement within the quick time period, it’s a restricted technique for changing minimal customers to long run individuals.
Trialling incentives as properly can include threat. A Zooniverse project, called Old Weather, trialled a aggressive rating technique to detrimental outcomes.