Advanced diagnostic framework for analyzing student emotional and cognitive patterns: A big data approach to personalized teaching

Abstract

In the realm of contemporary education, the emergence of big data technology has underscored the significance of analyzing student emotional and cognitive patterns, a process crucial for unveiling individual traits and needs throughout the learning experience. This study introduces a novel diagnostic framework, leveraging big data to precisely evaluate these patterns and enhance the allocation of personalized teaching resources. Traditional analysis methodologies, often limited by their inability to effectively process complex data structures and provide operational utility, are thus addressed. The proposed framework incorporates an integrative approach for emotional and cognitive diagnosis, employing Bayesian networks’ probabilistic distribution methods. This approach not only augments the interpretability of diagnostic outcomes but also streamlines the estimation of student ability distributions. Additionally, a groundbreaking multi-objective optimization strategy, grounded in the analytic hierarchy process (AHP), is presented, complemented by a solution method devised using a hybrid adaptive ant colony algorithm. This innovative method facilitates the efficient allocation of personalized learning resources, thereby significantly enhancing the adaptability of teaching content to individual student needs. The findings from this methodological and empirical investigation contribute a pioneering perspective to the field of personalized teaching. They provide educational practitioners with effective decision-support tools and offer researchers novel insights and methodologies in related domains.