The Design And Realization Of The Education Assistant System Of Merging MOOC Learning Behavior

With the rapid development of Internet technology,the online education of MOOC mode has been spread,which has been sought after by many learners,especially the educators in colleges and universities have begun to carry out the education practice of MOOC online learning and offline classroom integration.In the process of MOOC learning,a large amount of user behavior data is produced,which can objectively reflect the actual situation of learners.Therefore,the behavior data mining of MOOC has become the research hotspot in the field of education.The focus of this paper is to excavate the practical application value of MOOC learning behavior data.Firstly,the improved decision tree TG-C4.5 algorithm is used to mine the learning behavior data to achieve the goal of predicting and classifying the scores;Secondly,to solve the problem that educators who lack of data mining knowledge understand the results of mining,the algorithm integration is applied to the education assistant system to realize the visualization of the results of prediction analysis,and provide the basis for teachers to make decisions on the information of teaching early warning.The main research work of this paper is as follows:(1)The improvement of C4.5 algorithm based on traditional decision tree.In the traditional decision tree C4.5 algorithm,the time consuming and accuracy of the algorithm need to be improved,and the TG-C4.5 algorithm is improved by adding Taylor series and GINI index two methods,and the validity of the algorithm is validated by experiments on the UCI dataset.(2)Data acquisition and preprocessing.We adopt a university students on the MOOC behavior log,using Java technology related to parse the MOOC related log and cleaned and implement the data pretreatment,improved algorithm and select after preprocessing the data as the source of data for practical application.(3)Improved algorithm in MOOC application.For the data source obtained from the preprocessing,the Pearson coefficient and the information gain are used for attributes selection.Then the related algorithms are compared with the proposed TG-C4.5 algorithm through the attributes-selected MOOC dataset to verify the.validity of the practical application of the TG-C4.5 algorithm,which paves the way for the realization of the system in the following text.(4)The design and realization of the education assistant system which integrates MOOC learning behavior.Based on the basic function requirements of HTML,Flask framework and Python,the system integrates the proposed TG-C4.5 algorithm into the education assistant system,realizes the prediction of MOOC learning behavior achievement and the visualization of prediction results,and provides the teachers and learners with good early warning information and other decision support.