基于决策树的网络学院学生学业影响因素研究

发布时间：2018-03-03 22:13

本文选题：网络教育　切入点：网络学院　出处：《河南大学》2013年硕士论文　论文类型：学位论文

【摘要】：自1999年教育部开始大力兴办远程教育试点以来，网络学院在发挥他积极作用的同时也暴露出一些不足：由于网络教育学习者来源复杂，环境不固定，会造成精力投入的不够和学习时间不能保证等问题，以至于辍学、滞学等未完成学业的现象成为网络教育者面临的一个严重问题。目前关于网络教育学业影响因素的研究仍处于教育统计的层面，许多研究仍然建立在体会和感想等较低层次的基础上；运用的实证研究方法不够科学规范，如调查问卷的设计、抽样的代表性，数据的统计推理等方面不够严格，缺乏本土化的辍学理论的建构和理论上的提升。本文从几个方面对网络学院未完成学业的情况进行分析并得出易于决策的模型:首先，明确本研究的研究背景、研究现状、研究目的、论文的结构等；其次，简述我国网络高等教育的发展现状，并通过文献调研和理论分析，，了解“成人网络学习”、“数据挖掘”和“决策树”等相关概念和技术；接着改进了Kember模型，提出了适应我国特点的“网络教育学业完成影响因素模型”，并以此设计调查问卷，得出学生学习过程的相关信息；最后结合远程学习系统中的信息组成我们要研究的数据对象，引入决策树技术，利用Weka数据挖掘软件进行分析，找出每个因素对网络学习者完成学业的影响程度，并利用分析结果对可能有辍学危险的在读者和新入学者实施相应对策，有目的给予指导，提高预警转化率。本课题的主要任务是根据研究所在地的某网络学院的学生网络学习行为等信息，利用数据挖掘技术分析他们与学业完成之间的关系，得出一种最优的基于学业影响因素的决策树规则。通过数据分析得出最值得我们警惕的影响网络教育学业完成的因素，研究所形成的数据分析模型，使教育决策者对在读学习者和新入学者的干预管理提供了一个有力的手段，对提高网络高等教育质量有重大的意义。
[Abstract]:Since 1999, when the Ministry of Education began to set up a distance education pilot project, the network college has brought into play its positive role and also exposed some shortcomings: because of the complexity of the sources of online education learners, the environment is not fixed. It can cause problems such as lack of effort and lack of assurance of study time, so that they drop out of school. The phenomenon of uncompleted education, such as school delay, has become a serious problem for network educators. At present, the research on the influencing factors of online education is still at the level of education statistics. Many studies are still based on lower levels of experience and perception, and the empirical research methods used are not scientific enough, such as the design of questionnaires, the representativeness of sampling, the statistical reasoning of data, etc. Lack of localization of the theory of school dropout construction and theoretical promotion. In this paper, we analyze the uncompleted study of network college from several aspects and get the model which is easy to make decision: first, make clear the research background, research actuality, research purpose, structure of the paper, etc.; second, make clear the research background, the research purpose, the structure of the thesis, etc. This paper briefly introduces the current situation of the development of network higher education in our country, and through literature investigation and theoretical analysis, understands the related concepts and technologies, such as "adult network learning", "data mining" and "decision tree", and then improves the Kember model. This paper puts forward a "model of influencing factors of academic achievement of network education" adapted to the characteristics of our country, and designs a questionnaire to obtain the relevant information of the students' learning process. Finally, combining the information in the distance learning system, it constitutes the data object that we want to study. The decision tree technology is introduced, and the Weka data mining software is used to analyze the influence of each factor on the completion of the study. The results of the analysis are used to implement the corresponding countermeasures to the readers and new scholars who may drop out of school. Aim to give guidance to improve the early warning conversion rate. The main task of this paper is to analyze the relationship between students and their academic completion by using data mining technology according to the information of students' online learning behavior in a certain network college where they are located. Get an optimal decision tree rule based on academic factors. Through data analysis, we can get the most alarming factors that affect the academic completion of network education, the data analysis model formed by the research, It provides a powerful means for educational decision makers to intervene in the management of learners in reading and new scholars. It is of great significance to improve the quality of network higher education.
【学位授予单位】：河南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：G434

【参考文献】