In the post-epidemic era,online learning becomes routinized.Deep knowledge interactions among learners in online learning communities can trigger deeper online learning and improve learning efficiency.However,as a large number of studies have shown that knowledge interaction in online learning communities is not as ideal as expected,facilitating deep knowledge interaction among learners in online learning communities is now urgent to be addressed.Learning analytics can support learners to engage in deep knowledge interactions.Therefore,based on sufficient literature research,this study adopts a combined qualitative and quantitative research paradigm,by using a combination of literature analysis,questionnaire survey,structural equation modeling,and data mining methods to investigate the influencing factors of deep knowledge interaction of learners in online learning communities.Simultaneously,it combines the influencing factors to construct an intervention model of learners’ deep knowledge interaction based on learning analytics,with the aim of facilitating deep knowledge interactions of learners.From the theory of planned behavior,this study,introducing four exogenous variables,namely,the sense of achievement,self-development,assistant influence,and platform support,constructs a theoretical model of the factors influencing deep knowledge interactions among learners in online learning communities and puts forward research hypotheses.Drawing on mature scales at home and abroad to form measurement items,this study utilizes the Questionnaire survey and issues questionnaires through the Wen Juanxing platform.SPSS and AMOS are used to test the reliability,validity,and model fit.Finally,a structural equation model is used for hypothesis testing to derive the factors that influence deep knowledge interaction among learners in online learning communities.From the study,it has shown the following findings: the sense of achievement and self-development significantly influence deep knowledge interaction attitude;the assistant influence greatly affects subjective norms;platform support inserts a vital role in the perceived behavioral control;deep knowledge interaction attitudes,subjective norms,and perceived behavioral control significantly influence deep knowledge interaction intention,and deep knowledge interaction behavior is affected heavily by deep knowledge interaction intention and perceived behavioral control.Combining the factors influencing learners’ deep knowledge interactions in online learning communities,this study constructs a learning intervention model for learners’ deep knowledge interaction based on learning analytics,which can implement automatic interventions on the platform for learners and support teachers’ intervention.The model consists of three main modules: the intervention target diagnosis module,the cause diagnosis module,and the intervention module.The first one can diagnose learners who have not reached the level of deep knowledge interaction as intervention targets.Moreover,to automatically discern the level of learners’ knowledge interaction,this study designs an automatic interaction level assessment model based on the data from the course discussion forums of the 1st-4th opening sessions of "Civic Ethics and Legal Foundations" at Peking University and the 1st-8th opening sessions of "Civic Ethics and Rule of Law" at Wuhan University on the China University MOOC platform.The second module aims to diagnose the reasons why the intervention targets fail to carry out deep knowledge interaction.The set of measurement indicators and the corresponding neural network model are constructed for the reasons requiring platform data diagnosis and the corresponding questionnaire is designed for the reasons requiring questionnaire diagnosis.The last module mainly constructs a library of intervention strategies,designing an intervention method that combines the artificial one by assistants and the automatic one by the platform.Besides,this study applies the intervention model to the course "Fundamentals of Moral Cultivation and Law" of J University on the Learning Pass platform and verifies its effectiveness. |