基于Weka平台的网络教学数据分析研究与实践
发布时间：2018-08-24 19:30
【摘要】：现阶段,社会飞速发展,计算机技术不断革新。在教育领域,互联网+教育这一传统行业成为一个新的热点和蓝海,相关交叉学科的专家根据这一课题展开了广泛的研究。以共享知识资源为代表的在线课堂也发展的越来越快,譬如国内的清华在线,超星泛雅平台,以及国外的三大在线课程平台Coursera、Udacity和edX等课程资源极其丰富,一些优秀的教师也加入其中,带来了优质的师资资源。但是由于平台功能的限制,对于一些学生用户的学习日志、学习路径等行为数据没有相关的数据分析,无法对个体建立偏好模型,教师很难了解每一名学生的学习能力以及学习风格,因此不能针对性的制定个性化教学目标。由此,本文借助数据挖掘工具Weka,针对山东师范大学网络学习平台产生的大量教学数据、学生成绩数据、学生学习日志等,展开具体的挖掘分析。具体研究目标如下:1、利用关联算法,找到真正影响学生成绩的因素,给教师提供教学质量的分析和改进。2、利用相关的聚类分类算法分析学生学习能力,将具有相同或者相似风格的学生组合到一起,统一进行教学目标管理,发现学生与学生之间的联系。3、通过教学实践以及与老师之间的交流反馈,提出了四种学习风格数据量化指标,教师可以针对学生线上学习的数据区分不同学习风格的学生,结合聚类分析的结果,进行相关的任务布置和管理,实现个性化教学。4、针对不同教师的数据挖掘需求,提出了一个成绩分析平台模块框架,可以帮助教师针对学生的实际情况进行相关数据的挖掘,降低教师的教学成本。本文主要有两个创新点:1、技术上的创新:改进以往教育领域通过调查问卷来进行教学分析的做法,充分利用算法的优势,进行数据量化,通过科学的算法来对数据进行处理和分析,使得整个数据分析严谨性、操作性比较强。2、模型上的创新:建立了数据挖掘模型,通过学生数据分析不同学习风格的学生特点以及相关的量化指标,教师可以建立不同学习风格的学生分组,从而制定个性化教学目标,提高学生的创新意识和合作意识。通过网络教学平台的数据分析,教师可以根据不同学生的风格针对性的制定教学计划,学生可以变被动接受新知识为主动学习新的学习资源,教师实现因材施教。
[Abstract]:At present, with the rapid development of society, computer technology is constantly innovating. In the field of education, the traditional industry of Internet education has become a new hot spot and blue sea. The online classroom, represented by shared knowledge resources, is also developing more and more rapidly. For example, Tsinghua online in China, Chaoxing Pan-ya platform, and three online courses platforms, Coursera,Udacity and edX, are extremely rich in resources. Some excellent teachers also joined in, bringing high-quality teacher resources. However, due to the limitation of platform function, there is no relevant data analysis for some student users' behavior data, such as learning log and learning path, so it is impossible to establish preference model for individuals. It is difficult for teachers to understand each student's learning ability and learning style. Based on the data mining tool Weka, this paper analyzes a large number of teaching data, student score data and student learning log generated by the network learning platform of Shandong normal University. The specific research objectives are as follows: 1. By using the correlation algorithm, we can find out the factors that really affect the students' achievement, and provide teachers with the analysis and improvement of the teaching quality. Second, we can use the related clustering classification algorithm to analyze the students' learning ability. Students with the same or similar styles together, unified teaching objectives management, found the relationship between students and students. 3, through teaching practice and exchange feedback with teachers, Four data quantification indexes of learning styles are put forward. Teachers can distinguish students with different learning styles according to the data of students' online learning. Combining the results of cluster analysis, teachers can arrange and manage relevant tasks. To realize individualized teaching. 4. According to different teachers' demand of data mining, this paper puts forward a module framework of achievement analysis platform, which can help teachers to mine relevant data according to students' actual situation and reduce teachers' teaching cost. There are two main innovations in this paper: one is technical innovation: to improve the past teaching analysis in the field of education through questionnaires, to make full use of the advantages of the algorithm, and to quantify the data. Through the scientific algorithm to process and analyze the data, make the whole data analysis rigorous, maneuverability relatively strong. 2. The innovation of the model: set up the data mining model, Through student data analysis of students' characteristics of different learning styles and related quantitative indicators, teachers can set up groups of students with different learning styles, so as to formulate individualized teaching objectives and improve students' consciousness of innovation and cooperation. Through the data analysis of the network teaching platform, teachers can make teaching plans according to different students' styles, students can change passive acceptance of new knowledge into active learning resources, and teachers can teach students according to their aptitude.
【学位授予单位】：山东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13
本文编号：2201807
[Abstract]:At present, with the rapid development of society, computer technology is constantly innovating. In the field of education, the traditional industry of Internet education has become a new hot spot and blue sea. The online classroom, represented by shared knowledge resources, is also developing more and more rapidly. For example, Tsinghua online in China, Chaoxing Pan-ya platform, and three online courses platforms, Coursera,Udacity and edX, are extremely rich in resources. Some excellent teachers also joined in, bringing high-quality teacher resources. However, due to the limitation of platform function, there is no relevant data analysis for some student users' behavior data, such as learning log and learning path, so it is impossible to establish preference model for individuals. It is difficult for teachers to understand each student's learning ability and learning style. Based on the data mining tool Weka, this paper analyzes a large number of teaching data, student score data and student learning log generated by the network learning platform of Shandong normal University. The specific research objectives are as follows: 1. By using the correlation algorithm, we can find out the factors that really affect the students' achievement, and provide teachers with the analysis and improvement of the teaching quality. Second, we can use the related clustering classification algorithm to analyze the students' learning ability. Students with the same or similar styles together, unified teaching objectives management, found the relationship between students and students. 3, through teaching practice and exchange feedback with teachers, Four data quantification indexes of learning styles are put forward. Teachers can distinguish students with different learning styles according to the data of students' online learning. Combining the results of cluster analysis, teachers can arrange and manage relevant tasks. To realize individualized teaching. 4. According to different teachers' demand of data mining, this paper puts forward a module framework of achievement analysis platform, which can help teachers to mine relevant data according to students' actual situation and reduce teachers' teaching cost. There are two main innovations in this paper: one is technical innovation: to improve the past teaching analysis in the field of education through questionnaires, to make full use of the advantages of the algorithm, and to quantify the data. Through the scientific algorithm to process and analyze the data, make the whole data analysis rigorous, maneuverability relatively strong. 2. The innovation of the model: set up the data mining model, Through student data analysis of students' characteristics of different learning styles and related quantitative indicators, teachers can set up groups of students with different learning styles, so as to formulate individualized teaching objectives and improve students' consciousness of innovation and cooperation. Through the data analysis of the network teaching platform, teachers can make teaching plans according to different students' styles, students can change passive acceptance of new knowledge into active learning resources, and teachers can teach students according to their aptitude.
【学位授予单位】：山东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13
【参考文献】
相关期刊论文 前10条
1 张明阳;常盼;徐冬冬;顾芝亚;王东林;陶陆阳;;独立学院医学生法医学个性化教学模式的探究[J];教育教学论坛;2016年52期
2 包耕;张玲乐;;关联规则隐藏算法综述[J];软件导刊;2016年11期
3 孙崴;刘学敏;许红梅;;基于大学生认知风格特征的MOOCs课程建设研究[J];现代教育科学;2016年10期
4 杜唐;徐仕宝;李明东;;基于聚类分析算法的应用性改进[J];信息通信;2016年10期
5 陈志飞;冯钧;;一种基于Apriori算法的优化挖掘算法[J];计算机与现代化;2016年09期
6 徐刘杰;;基于网络学习平台的翻转课堂教学研究[J];软件导刊(教育技术);2016年07期
7 应国良;;国际学校中“个性化教学”带来的启示——评《如何进行个性化教学:来自国际学校的启示》[J];中国教育学刊;2016年06期
8 陈俞;赵素云;陈红;李翠平;孙辉;;统计粗糙集[J];软件学报;2016年07期
9 李牧南;;基于关联规则挖掘竞争情报研究前沿分析[J];情报杂志;2016年03期
10 王景中;张存正;;用于网络行为分析的一种改进K-means算法[J];北方工业大学学报;2016年01期
相关硕士学位论文 前3条
1 鲍素贞;数据挖掘技术在个性化网络教学平台中的应用研究[D];聊城大学;2015年
2 滕子牧;数据挖掘关联规则算法研究与应用[D];辽宁科技大学;2015年
3 高鹏;基于数据挖掘的个性化网络教学平台研究[D];西北大学;2005年
，本文编号：2201807
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2201807.html
