An Empirical Study on Learning Path Optimization by Data Collection and Computational Analysis of English Learners’ Behavior in the Era of Big Data

Learning path optimization aims to generate and optimize a knowledge learning sequence for learners that best meets their knowledge needs.This study focuses on the important role of online learner behavior in personalized path planning.By constructing a knowledge point difficulty model and a learning behavior prediction model based on online learning behavior, together with a user-based collaborative filtering recommendation algorithm, a personalized learning path is proposed comprehensively.The MOOC websites "College English 1" and "Xuedang Online" are selected as sample data to analyze the online learning behavior of English learners and verify the learning effect of the learning path proposed in the article through the change of students' online time.The personalized teaching model based on the learning path is investigated in practice by taking the college English course in school A as an example.Compared with the traditional teaching mode, the optimized learning path shows a significant difference of 0.01% in the dimensions of learners' "knowledge and skills", "process and method" and "affective attitude".The mean values of the optimized blended teaching mode are 4.12, 4.33 and 4.07 respectively, which are all better than the traditional teaching mode.It shows that the English learning path proposed in this paper is conducive to enhancing students' personalized learning needs and provides a reference for promoting the effective implementation of personalized learning in the information technology environment.