Analysis And Prediction Of Learning Behavior Of Big Data In Online Education Platform

With the rapid development of Internet technology,human beings are gradually stepping into a new era,information technology has penetrated into all aspects of our lives.Education has always been a hot issue of great concern in China.In combination with the new needs of modern social development and the high efficiency and convenience of the Internet,the perfect combination of education industry and Internet technology is playing an irreplaceable role in our study and life.The learning mode has been transformed from traditional teachers and students in the same classroom to horizontal through direct contact A new online learning mode across time,space and place.The public has a high degree of acceptance of this new learning mode,and the online education and learning mode is more and more widely accepted,and the people are more and more fond of the development mode of online education.In the face of the booming development of new forms of social education,how to evaluate online education learning behavior and learning effect is particularly important.In the new situation,once the users of online education platforms choose the way of online learning,they will generate a wealth of learning behavior data in the process.If these data are forgotten in the corner and covered with dust,then they will lose the value of their data.We need to explore these data,explore the laws behind these data,study the behavior data records of online education platform learners,and provide effective suggestions for online education platform decision makers to create a better online learning environment and platform rules.At the same time,the International Conference on education data mining emphasizes the importance of big data development and application in education industry.At the same time,due to the wide application of artificial intelligence technology in various fields,for the development of education,the application research of data mining technology in this field at home and abroad is more urgent,and data mining technology has become one of the main forces in the development of education big data.Nowadays,the research on the prediction of students’ situation in various aspects by using data mining technology in Colleges and universities at home and abroad has become increasingly mature.Researchers use different data mining algorithms to preprocess the data of individual behavior characteristics of different types of students from different perspectives,and put forward different research results on the prediction of various aspects of online education.Based on the importance of new forms of Education under the Internet in today’s society,online education platform is rising day by day,gradually becoming a rising star in the new business sector of the country.The online education platform has gradually enriched big data resources.In order to better understand the user behavior of online education platform,explore the hidden information behind education big data,explore the relationship between education data information and learners’ behavior and education effectiveness,so as to improve people’s awareness of online education platform and guide online education related decisions more directionally To promote the further development of online education platform.Based on the research on the learning behavior of online education learners in domestic and foreign academic literature,this paper uses the open data set of canvas network as the data source to summarize and analyze the learning behavior of the online education platform,and uses the descriptive statistical method to describe the canvas in detail The basic information of network platform learners,course setting information and the distribution of learners’ types.Online education platform should not only adopt different guiding education methods for different types and characteristics of learners,establish more targeted platform rules,attract groups with different social characteristics to participate in,but also reasonably set up online education content and modules,so that online education field can achieve real widespread popularization.Then,Xgboost algorithm,a classification method in data mining technology,is used to analyze and predict the factors that affect the learning performance of online education platform learners.By building a performance prediction model,the learning performance prediction of online education platform learners is successfully realized,and then the online education industry leaders are more directional to make decisions in this field.At the same time,this paper ranks the importance of influencing factors to provide new ideas for the future research in the field of online education.This paper finds that the learning behavior of learners on the education platform is the main factor affecting their performance.Therefore,online education industry should pay more attention to the enthusiasm of learners for platform activities,guide learners to actively communicate with each other,improve learning efficiency,increase the practicability of online education,and actively play the positive effect and value of online education in the society.The open data set of canvas network selected in this paper has some data shortage.Therefore,the integrated classifier model Xgboost is selected to build the learning achievement prediction model of online platform learners,Data mining is a complex and interesting science,and the choice and accuracy of its methods in the study of online education learners’ performance prediction need to be further improved.At the same time,it also needs the improvement of the development and application of online education big data.