Prescribing Deep Attentive Score Prediction Attracts Improved Student Engagement

This paper provides a glimpse of how a more accurate prediction of a learner’s academic performance could influence the student’s engagement with an online platform intelligent tutoring system(ITS). The study was based on the assumption that accurate score prediction models would result in an increase in overall activity with an ITS.

Definition:

The score prediction model predicts how a learner would perform in a standardized test based on the learner’s responses to questions.

Definition:

An intelligent tutoring system(ITS) is a computer-aided learning environment designed to provide immediate and customized instructions that are adaptive to a learner’s knowledge.

A/B tests were conducted to confirm our hypothesis, and results indicate that both revenue, retention rate, and average revenue per paid user(ARPPU) have increased with a better version of our score prediction model.