A Study of Online Behavioural Data Analysis and Teaching Intervention Strategies for College English Learners

Abstract This study aims to understand the characteristics of online learning behaviors among modern online education learners, implement these behaviors, propose strategies for learning support services, and propose reform proposals to enhance the quality of online teaching. This paper collects English learners’ online learning behavior data from relevant learning platforms and quantifies as well as processes the data. Subsequently, a model of behavioral analysis is constructed using social network analysis and the DBSCAN clustering method. The teacher takes the lead in teaching social behavior, ensuring that other English learners do not feel isolated from each other. By extracting the behavioral data of 5163 students for analysis, the English learners’ learning discussion time is mainly several at the beginning of the semester and the end of the semester, and the number of discussions at other times is almost zero. The DBSCAN clustering algorithm was used to classify English learners into three types, and intervention strategies were proposed for each type to improve teaching quality.