Research And Implementation Of Online Learning Behavior Data Analysis And Visualization Based On Hadoop

With the rapid development of information technology and the Internet,great changes have taken place in people's life and study.Against this background,online learning has risen rapidly and become an important part of modern education.Unlike traditional education,online learners are no longer limited by time and place,and they can study at any time.However,it is precisely because teachers and learners are separated in time and space during online learning that it is difficult for teachers to observe directly the behaviors of learners in classroom,timely find problems and provide effective guidance.In order to solve the problems,it is necessary to collect massive data recorded in online learning platform,and find the valuable information.This paper combines Hadoop and the traditional method of data collection to collect the data.By cleaning and preprocessing noisy or missing data,the paper realizes a comprehensive analysis of the behavior of learners and facilitates the professional management of teachers and platform managers.The main work and contributions of the thesis are described as follows:1.By illustrating the research characteristics and research status of online learning behavior analysis at home and abroad,and combining with the educational principles related to online learning behavior of learners,this paper makes these lay a theoretical foundation for the analysis of behavior data;Introducing the data acquisition and key technologies of visualization is used to provide technical support.2.Propose and construct an online learning behavior data analysis and visualization model.First,the factors that influence learners' online learning behaviors are analyzed,and targeted behavior parameters are selected.Based on this,the overall structure of online learning behaviors data analysis and visualization is constructed,including the data acquisition,preprocessing,analysis,and visualization;In addition,the content involved in the four parts of the model is studied.3.The model of online learning behavior data analysis and visualization is studied.The big data processing technology based on Hadoop is used to collect and preprocess the data;This paper focuses on the learning behavior of accessing course,watching video,and discussion and the data is analyzed from three aspects,including the descriptive analysis of learners'basic learning behaviors,the correlation analysis of learning effects and the clustering analysis of the learning behavior based on K-means algorithm;At the same time,the results of the experimental analysis are visually presented,and the enlightenment for learners,teachers,and platform managers is obtained.This paper combines the theory of behaviour science with the big data processing technology to build an online learning behavior data analysis and visualization model.Through in-depth analysis of the massive learning behavior data,it helps to comprehensively analyze the learning behavior of learners,facilitate professional teaching management of teachers,optimize the construction of the learning platform,and promote the development of online education.