Research On Interactive Behavior Of SPOC Forum Based On Learning Analysis Technology

SPOC,which effectively combines the advantages of MOOCs and traditional courses,has made great achievements in online education.This special mode of campus education has also entered higher education,and its development is in the ascendant.As an important part of web-based learning space,course forum plays an important role in students' online collaborative learning.However,as far as the current situation is concerned,the interactive behavior in course forum is not ideal,but studies have shown that some factors in forum interactive behavior are closely related to students' academic performance.Therefore,how to make full use of the existing data scientifically,find facts and draw conclusions from the data,and then take corresponding strategies and measures to improve the current situation of interaction,thus improving the quality of education,has become an urgent problem to be solved by educators.Learning analysis technology can discover potential problems of students,evaluate their academic progress and predict their academic performance by analyzing the massive data of students.When learning analysis technology is applied to the research of forum interactive behavior,valuable information can be found from a large amount of data,which is helpful to understand and optimize learning and learning environment.The research is based on the forum interactive behavior data in the SPOC course "C Language Programming" of Sichuan Normal University in spring 2017.Firstly,the social network analysis in the learning analysis technology is used to analyze the characteristics of the interactive network,aiming at mastering the interaction of interactive network,network members and their communities from three aspects: whole,individual and group.At the same time,interactive network indicators which may have a positive correlation with academic performance are calculated.Secondly,the content analysis in the learning analysis technology is used to analyze the characteristics of the interactive content,to mine and judge the students' involvement in the discussion of specific topics.At the same time,the interactive content indicator of knowledge construction level is calculated.Thirdly,interactive network indicators and the interactive content indicator are combined with the existing literature to construct interactive behavior indicators of SPOC forum.Then,the statistical analysis in the learning analysis technology is used to explore the relationship between each interactive behavior indicator and academic performance one by one,and the combination of interactive behavior indicators with the greatest explanation of academic performance is identified.In addition,in order to explore the possibility of early warning of academic performance,the forum interactive behavior data are divided into four different periods by month,then we compare and analyze the performance differences of the prediction models established in different periods on students' SPOC grades and final grades,and find out the best time to carry out early academic performance warning.Finally,taking into account the problems reflected in the above analysis,we develop corresponding measures and strategies for SPOC forum interactive activities along four main lines: improving the interactive form,improving the interactive content,improving the evaluation method and optimizing the platform function.