Pattern Recognition and Intelligent Analysis of Large-Scale Student Behavior Data in the Perspective of Educational Technology

This study attempts to mine students’ learning behavioral patterns and provide teachers with suggested interventions. This study takes the online behavioral data of large-scale learners as the research object, mines and extracts their learning behavioral features, then divides the extracted behavioral features into 11 specific behavioral indicators, and uses the K-Means algorithm to perform cluster analysis on the learning behavior of learners, and then compares the different learning groups in terms of learning motivation, time investment, learning effectiveness, and learning interactions in four directions, respectively. The lagged sequence analysis was used to explore the differences in the learning behavior sequences of different learning groups. The differences in the total time spent on the task and the number of replies are significant among the four categories of learners: general, negative, interactive, and active. The “average learners” had the best learning outcomes with a score center greater than 88, and the “negative learners” performed poorly in all four learning behaviors. Teachers can improve students’ learning motivation, interactive behavior, and time investment to enhance their performance on different learning behaviors while ensuring that the frequency of students’ behaviors is evenly distributed. The intelligent analysis method presented in this paper provides a reliable and reasonable basis for teachers’ teaching interventions.