Prediction And Intervention Of Learning Effect Based On Online Learning Behavior

MOOC promotes the spurt development of online education,but online education faces problems such as low participation,low input and high dropout rate.Most current studies analyze educational data through various technical tools to alleviate the online learning crisis.However,these studies favor the instrumental rationality of learning analysis and neglect the value rationality based on human development.Therefore,exploring the influencing factors of online learning effect and formulating interventions that reflect its value rationality are the key to improving the learning quality.This thesis constructs a theoretical model of the influence of online learning effect based on learning input,which is preliminarily verified by questionnaire survey and structural equation model,and further verified by online learning behavior data.The results show that learning input has a significant and positive impact on online learning effect.Subsequently,based on the key variables affecting online learning effect,multiple prediction models are used to fit the relationship between online learning effect and variables,and selected the optimal prediction model with accuracy and comprehensive evaluation indicators,so as to predict the learning effect,identify potential risks and classify learning groups,and formulate personalized intervention strategies for different learning groups based on learning input.Finally,a learning system to enhance online learning effect is designed and developed,which includes six modules: login registration,autonomous learning,discussion,homework,learning analysis and intervention.By designing intervention strategies for different learning groups,this study aims to teach students in accordance with their aptitude,thus demonstrating the value rationality of learning analysis.