A Smart Learning Assistant to Promote Learning Outcomes in a Programming Course

Abstract

Blended learning has gained wide popularity, but its superiority is limited by insufficient connection between online and offline learning due to technological anxiety and complexity, which hampers the achievement of prospective learning effect. To shatter these limits, a smart learning assistant based on Wechat Mini Program is proposed that incorporates a score ranking mechanism based on explainable machine learning to improve learning interests in programming, a learning material recommendation with deep neural networks to solve the student's confusion in personalized learning source selection, and a learning review mechanism based on deep learning achievements to enhance teacher-student communication and student-student cooperation in learning. In addition, approximately 3200 learners are involved to investigate learning requirements and test system performance. The experimental and practical results demonstrate the superiority of the smart learning assistant and the effectiveness gained by promoting learning outcomes in blended learning.