Construction and Algorithm Research of Online Learning Interaction Behavior Model Based on FBM

Abstract

Abstract The development of online learning environments has facilitated changes in educational interactions. This study focuses on an in-depth exploration of learners’ online interaction behaviors based on the theory of the Fogg Behavior Model, all in the context of accurately depicting and optimizing these interactions. Considering the complexity and richness of online education platform data, this paper proposes an improved analytical methodology to accurately model and predict learner behaviors, expecting to provide a scientific basis for the development of personalized teaching strategies and the enhancement of learning experience This methodology is based on the Fractional Brownian Motion and BAS algorithms to analyze the data of online learning platforms. The study collected data such as login, learning status, and course grades, and used the combination algorithm to optimize the Hurst index estimation and improve prediction accuracy. The number of view notifications is significantly correlated with the interactive learning behavior of resources at the 0.01 level, and the correlation coefficients are 0.297, 0.557 and 0.360. The algorithm accurately reflects learner behavior, provides optimization strategies for online learning platforms, and enhances learning efficiency and interaction quality.