Application Of Bayes Method To Personalized Learning In Internet Education

“Making no social distinctions in teaching” and “teaching students according to their aptitude” are the goals that the education industry has been pursuing for a long time.How to cover more groups with higher quality education resources at lower cost and provide more targeted and practical teaching programs for different groups is the cause that education practitioners have been striving for.After decades of development,people's expectation for Internet education has changed from the informationization of educational resources and the equalization of resource sharing to the high quality of educational resources and the intelligence of teaching efficiency.Education practitioners and Internet practitioners are constantly studying and spending time.Internet technology not only solves the problem of educational resources sharing,but also further improves the quality of educational resources and teaching,so that students can get better teaching resources and learn more efficiently and more closely to their own situations.The Bayesian network model is a type of network model method that can be used to express various uncertainties and scientific knowledge.Because of its model characteristics that it becomes a representative data mining and analysis tool.In recent years,with the upsurge of artificial intelligence,Bayesian network has become the object of research by many scholars.Among them,in the Internet teaching,because the learning situation of each student is very different,it is very difficult to develop a corresponding learning plan for each student,but the characteristics of the Bayesian network can accurately understand each student's own situation And carry out corresponding scientific analysis,and then classify and evaluate it,and finally recommend personalized learning methods based on the results obtained in the above steps.This article focuses on the following three aspects of the learning process and evaluation methods of modern Internet Education and learning:(1)Aiming at the learning and assessment of the courseware in the course,a method to count the passing rate of the courseware and the knowledge associationbetween the courseware based on the course organization and evaluation results is proposed.First,establish the corresponding statistical model by organizing three main evaluation methods: pre-class comprehensive test,pre-school quiz and post-school quiz,and based on Bayesian formula to infer online the most suitable courseware for students.The analysis of the experimental results shows that the model can be used to adjust the effective push degree of the next courseware online in real time and improve the user's learning efficiency.(2)Aiming at the evaluation and comparison between different courses,a method based on Bayesian to evaluate the learning ability of courses is proposed.First,the user's learning ability score is estimated based on the historical evaluation results of the course and the user's pre-class comprehensive test;Then,in the user learning,according to the results of the first test and the post class test,the student's learning ability score is modified in real time on the premise of Bayesian formula.It is known from the experimental results that the introduction of Bayesian network can effectively improve the accuracy of evaluating the learning ability of courses.(3)The two methods are combined to design a personalized learning module that can improve the user's learning efficiency and curriculum learning evaluation method.Collect courseware learning data and evaluation index statistical data based on the first two methods,flexibly choose different compliance evaluation schemes for the collected data,or judge the theoretical compliance line for the collected data based on prior knowledge,and then based on the data.The characteristics of the analysis and adjustment of the standard line,and finally get the ideal evaluation plan.This article uses tools to create and evaluate Bayesian-based personalized learning modules.The research results show that the personalized learning module proposed in this paper can continuously revise and perfect the best path for the user to complete the course last time according to the real-time situation.