Research On Self-regulated Learning Path Mining And Promoting Strategies In Inquiry Based Learning

With the improvement of educational informatization and the integration of education and learning analysis technology,the recording,monitoring,and quantification of self-regulated learning(SRL)processes,which play a crucial role in exploratory learning,have become increasingly convenient.Therefore,more and more researchers are exploring ways to improve learners’ learning performance and innovative practical abilities by exploring and analyzing the regulatory behaviors and behavioral sequences of self-regulated learning in exploratory learning.At present,research on self-regulated learning in exploratory learning mainly focuses on three dimensions: learner classification,sequential pattern mining,and promotion strategies.Although there have been good achievements in related research,there are still limitations such as fuzzy learner classification,lack of in-depth learning path mining,and lack of systematic promotion strategies.Therefore,this study proposes solutions to address the shortcomings of existing research,and the specific research content includes the following three aspects:1.Self regulated learning data acquisition and learner analysis in exploratory learning.This study uses the learning behavior data generated by learners conducting exploratory learning on online simulation platforms,and then encodes the learning behavior based on the self regulatory learning behavior coding framework in exploratory learning.Then,through dimensionality reduction and clustering in machine learning methods,learners are divided into different clusters based on their learning behavior data,Finally,the learning cluster is defined by preferences in different self-regulated learning behaviors.To achieve a clear classification of learners,thereby gaining a better understanding of learners and achieving individualized teaching.2.Mining self-regulated learning paths in exploratory learning.This research determines the transfer probability of self-regulated learning behavior among various learners through Markov chain analysis,analyzes the transfer probability between high and low performance learners,and summarizes the factors that affect learner performance.By mining the learning behavior sequence of high to low performing learners among similar learners,we identified learning paths that have different impacts on learners.Combined with the results of interview analysis,we explored the explicit problems of low performing learners in self-regulated learning behavior and their implicit learning needs,providing important reference for improving learners’ self-regulated learning level.By delving deeper into the learning path of learners and gaining a comprehensive understanding of their learning process,we can effectively assist them.3.Strategies for promoting self-regulated learning in exploratory learning.This study identifies the common and individual needs of different types of learners based on their learning behavior data.Based on the learning needs of learners,a systematic promotion strategy for improving learning ability has been proposed,and the promotion strategy has been further refined into executable intervention measures.To systematically design promotion strategies and provide precise learning service support for learners with different learning processes and preferences.Research analysis found that:Firstly,based on the theories of exploratory learning and the coding framework of self-regulated learning both domestically and internationally,according to the actual situation of this exploratory experiment,the self-regulated learning behavior in exploratory learning is divided into six behaviors: task analysis(TA),planning(PL),refinement(RT),execution(EX),monitoring(MO),and reflection(RE).Based on the differences in different learning behaviors of learners in exploratory learning,machine learning clustering analysis is used toThere are three types of learners: 35(45.5%)are ’concise learners’,34(44.2%)are’practical learners’,and 8(10.3%)are ’active learners’.Second,through Markov chain transfer probability analysis and sequential pattern mining,supported by data,explore the differences between different types of high-and low performance learners.Combined with the analysis of the interview results of learners,among them,learners of "concise learners" need to have a clear understanding of the task and need to continue to try,but after task analysis,they can not blindly carry out the operation,and need to plan the implementation.Learners of ’practical learners’ need to review tasks and set goals appropriately after refinement,and it is best not to directly refine tasks after task analysis.Active learners will engage in frequent self-regulated learning behavior transitions during exploratory activities,but self-regulated learning behavior mainly shifts to task analysis and execution.Analyzing tasks and correcting their learning goals appropriately during the learning process can help complete learning tasks.Thirdly,based on the entire experimental research results,this study systematically proposes promotion strategies from three levels: activation of learning motivation before learning begins,optimization of personalized learning paths during the process,and evaluation feedback after learning ends.Conducting exploratory learning in undergraduate classrooms provides a scientific basis,and optimizing self-regulated learning in exploratory learning provides precise strategies and methods for improving learners’ performance.