Analysis And Research On Online Scientific Inquiry Behavior Based On Machine Learning

In recent years,with the gradual widespread application of internet big data and visual simulation in the field of teaching,scientific experimental teaching that used to require specific conditions and instruments can be completed through virtual laboratories.Virtual laboratories can flexibly and accurately set various experimental process parameters,as well as record the complete process data of students’ scientific exploration,providing sufficient data support for in-depth analysis of students’ learning habits and changes in learning level.On the other hand,existing research based on virtual laboratories has mostly focused on different types of exploration platforms,whether different exploration strategies and methods have significantly improved students’ scientific performance,scientific skills,and other aspects.There is a lack of using machine learning technology to analyze the fine-grained process data collected by the platform,making it difficult to come up with more personalized decision-making recommendations.Therefore,this study conducted experiments and collected data based on an independently developed online scientific exploration platform.Machine learning methods were used to mine the collected student learning process data at a fine-grained level,in order to identify key behavioral features that can predict and represent learners’ exploration status.Through the use of visual dashboard technology,key features and indicators were presented in real-time,providing process guidance and teaching intervention basis for teachers.The main work of this article is as follows:1.Based on the standard process of scientific exploration and related theories,transform the scientific concept knowledge corresponding to the exploration topic into process problems and embed them in the exploration process.Select experimental subjects with the same level of preparation,and design and implement scientific exploration experiments with theoretical basis.2.Collect the entire process data of students’ online scientific exploration and analyze the key factors that affect learning effectiveness.Using machine learning algorithms to perform fine-grained mining on it,identifying key behavioral indicators,laying a reliable foundation for the division of student groups and the recognition of high-frequency behaviors in different clusters.3.Design a visual dashboard for teachers and students.Design a visual dashboard with basic information and key behavioral indicators related to students and grades,providing scientific teachers with teaching references,reducing their workload,and providing scientific basis for early teaching interventions;Provide students with periodic exploration summaries and personalized feedback.