Development and Assessment of Recommendation System Based on Behavioral Intervention Analysis

Abstract

Computer science education in primary and secondary schools uses the. visual programming language "Scratch " to learn computational thinking. However, the limited number of course hours and overwhelmed teachers delay students’ learning progression. Although many Scratch instructional videos are available online, students still fail because of a lack of keyword-searching ability. Given this learning dilemma, we use the existing teaching videos online as hints assisting students in learning "Scratch." This project uses OpenCV to record the program blocks and analyze the behavioral intentions, and then compares and analyzes it with the video content to recommend and mark relevant segments that students need. We believe that through this recommendation system, intelligent hints in the form of videos can cultivate students’ self-learning ability and improve current teaching adversity.