Research on the construction of personalized learning system supported by big data in education

Abstract

In order to optimize the drawbacks of current personalized learning systems on the market, a big data algorithm is used to optimize the personalized learning system. This paper first analyzes the system model, constructs the basic framework of the system, and optimizes the algorithm based on a collaborative filtering algorithm that converts user behavior into ratings and recommends personalized learning content for learners. Since learning resources and learners are interconnected, this connection is analyzed by an ant colony algorithm to provide the optimal path for students to learn and create a personalized learning path. After comparing the student models, we understand that the clearer the description of students' interests, the clearer the accuracy returned, where User A and User C have the highest similarity of 99% and accuracy of 85% and 88% respectively, proving the feasibility of the system. The personalized learning system supported by big data in education can optimize the drawbacks of the current personalized learning system in the market, meet the concept of teaching according to student's abilities, and outperform the learning system in the market.