Data-driven Teaching Functions And Evaluation Of Learning Effects In Online Education

In the environment of educational informationization and education globalization,the high quality online education platform,represented by MOOC and Khan Academy,is one of the important ways to acquire knowledge.With the rapid increase in the number of online learners,every year,a large amount of learning content and education management data generated.At the same time,with the large data technology is gradually mature,large data used in the field of education has become a trend.how to effectively use online learning behavior data to improve learners learning efficiency,has become one of the challenges facing online education,Circumstances,personalized education and data-driven teaching will become the inevitable requirement of future online education.This thesis analyzes the impact of educational data on China's development strategy,summarizes the relevant policies,researches status at home and abroad,and points out the great significance and challenges of educational data on educational reform.It puts forward the data-driven teaching process framework based on the data collection and mining analysis of online education,and realizes the teaching progress reengineering and teaching content reengineering.Through the real data of the online education platform of a university,the data analysis and verification of the factors influencing the learning effect in the online learning process were carried out.The innovative work of this thesis are as follows:(1)This thesis designs the process framework of data-driven teaching,based on the research of emotion recognition and learning analysis,and learns the learning progress reengineering module and the learning content evaluation module,which is based on the learning effect evaluation,and realizes the content and learning scheme to generate personalized teaching aims.Through the basic data record in the process of registration,the operation records in the process of the platform,the behavior records in the learning process and the state data record analysis in the learning process.the learner's characteristic type analysis,the learning effect analysis,and finally the learner's learning Content optimization recommendations and learning programs are automatically generated.(2)With the aid of R language platform,the data of distance education platform in colleges and universities are analyzed and visualized,and the factors influencing learners' learning efficiency in online education are explored and verified.Based on the characteristics of statistical and association rules mining algorithm,the data are preprocessed and analyzed by statistic and Apriori association analysis algorithm,and the results are visualized.The analysis shows that the average score given by the teacher's assignment is negatively related to the amount of the teacher's instruction.The student's job is generally "delayed".The learning effect is positively related to the number of logins,online time and online discussion.Finally,through the analysis of the results,the thesis proves the feasibility of the behavior-driven teaching given the online learning process to improve student learning results of the proposal.and has some reference significance to the realization of the educational modernization goal of the 13th Five-Year Plan.