The Research On The Learner Knowledge State Modeling Integrating Academic Emotion And Its Application In Online Learning In Middle School

Learner modeling refers to the establishment of an abstract representation model of real learners and it' s one of the core technologies of adaptive learning systems.With the student characteristics simulated by the learner model,the adaptive learning system predicts the learner's learning state and automatically performs learning strategy adjustment,learning resource recommendation and learning early warning,and truly realizes the learner-centered.This study proposes a learner knowledge state model that integrates academic emotion;based on empirical research,it explores the effectiveness of the AFM+E model in improving online learning for middle school students.This study first uses literature analysis method to discuss the relationship between confusion and learning results based on the theories related to learning emotions.Furthermore,on the premise of given learner emotion,a knowledge state prediction model combining learning confusion emotion with Additive Factor Model is proposed by using mathematical modeling method.Finally,an empirical research scheme is designed and implemented.It is first applied to the data set of an mathematics online learning platform in a middle school,and then a data set is collected by a self-developed experimental platform for make a comparative research.we compare and discusse the new model with traditional models to verify the new model in the predictive performance in two data sets and the effectiveness of the new model in online learning applications for middle school students.Research findings:(1)confusion is a good characteristic factor to predict learning results.(2)confusion has situational limitations on the prediction of learning results.Only under the condition that learners learn effectively,the generation of confusion is positively correlated with learning results.(3)the learner knowledge state model that integrates learning emotions performs better than the existing model that only considers behavioral characteristics,and can predict learning results in Online learning in middle school more accurately.This research enriches the construction of learner models,explores the unique nature of the confusion,attempts to integrate learning behaviors and learning emotion characteristics in model construction,and provides more thinking about personalization learning for Secondary education.