The Data-driven Analysis And Research On Student Content Publishing Behavior In Network Learning Space

In recent years,the construction and application of network learning space is one of the emphasis in the development of educational informatization in China,but there are still difficulties in its high-quality construction and application in the stage of basic education.Paying attention to students' application of network learning space has positive significance to the construction and development of network learning space.Content publishing behavior is one of the important behaviors of students in the application of network learning space,but at present,there is still a lack of in-depth research and evidence support for the performance law of students' content publishing behavior.Based on the above background,this study is based on objective data in the network learning space,comprehensively using data mining technology,data visualization technology,development research method,mathematical statistics,etc.,to study the content publishing behavior of students,and analyze the performance the content publishing behavior of students in network learning space,including the following aspects:(1)Analyze the behavior characteristics of content publishing and build a behavior analysis modelThis research constructs a behavior analysis model which consists of data collection,data storage,data analysis,data visualization presentation and result analysis,and then uses data mining technology,data visualization technology,mathematical statistics and development research method to carry out specific research and design.(2)Data driven publishing behavior data analysis is realized by integrating various technical methodsThis research is based on the behavior analysis model,according to different analysis needs,using data mining technology,data visualization technology,mathematical statistics,development research method and other technologies and methods to realize the analysis of the data related to content publishing behavior of students in the network learning space.All kinds of analysis results provide the basis of further mining of behavior performance rules.(3)Analyze the characteristics of students' content publishing behavior in network learning spaceBased on the analysis results of content publishing behavior performance data,this study conducts a comprehensive analysis on the content publishing behavior performance of students in space from two aspects of behavior characteristics and content characteristics,and finds that:1)students log in more frequently in space,and have higher interest in using network learning space;2)students in different stages have greater differences in application level,and primary school students are more active;3)there are obvious differences in students' publishing behavior in different spaces;4)students'publishing content is generally lack of quality,and their participation and investment in learning tasks are not high.(4)Analyze the influencing factors of students' content publishing behavior in network learning spaceBased on the results of content publishing behavior performance analysis,this study further uses correlation analysis and other methods to mine the factors that affect the content publishing behavior of students,and finds that:1)the publishing behavior of resources is closely related to the type of resources and unrelated to the subject of resources;2)the publishing behavior of articles in personal space will affect the behavior of publishing pictures and comments;3)the publishing behavior of students in learning space is affected by specific learning tasks and students' active feedback on the completion of homework;4)the behavior characteristics of content publishing in space are related to spatial types.(5)Analyze the relationship between students' content publishing behavior and learning in network learning spaceBased on the dimensions of different publishing behaviors and significant characteristics,this study first analyzes the relationship between behavior and learning effect,and finds that:1)publishing behavior in learning space has a greater positive impact on learning than publishing behavior in non learning space;2)publishing articles and pictures in non learning space has a positive effect on learning effect.Then,based on the K-means clustering algorithm,this study get five types of publishing groups:outstanding,sharing missing,regular,comprehensive and passive,and analyze their publishing behavior and learning effect.The innovation of this study is reflected in:1)relying on large-scale network learning space,this study explores the relevant laws of content publishing behavior of primary and secondary school students in network learning space,and enriches the research on network learning space and content publishing behavior,and provides a certain reference for improving the construction and application of basic education network learning space.2)compared with the previous research on network learning space based on questionnaire survey data,this research is driven by the process data of network learning space,and comprehensively uses data mining technology,data visualization technology,mathematical statistics method and development research method to analyze the data,which not only reduces the subjective impact,but also mines potential laws from a large number of process data and has the analysis of content publishing behavior more accurate and comprehensive.