数据挖掘在大学英语教学和测评中的研究与应用

发布时间：2019-05-15 09:13

【摘要】：随着“互联网+”的出现,各行各业都发生了翻天覆地的变化,“互联网+教育”已经成为一种发展趋势,为大学英语教学和学习带来了新的机遇与挑战。通过在线英语学习平台进行教学和学习,运用数据挖掘技术对学生学习数据进行分析并建立相关模型,探索英语考试与各个要素之间的相关性,对学生学习和教师教学具有重要的价值。笔者对此进行了深入的研究。本课题首先描述了数据挖掘与大学英语教学中的国内外研究现状,并提出了本课题的研究内容与研究方法。其次,对本研究中的理论基础进行介绍,介绍了数据挖掘的功能、流程以及常用的机器学习模型。最后通过北京邮电大学本科生使用大学英语教学和测评系统产生的数据对学生进行构建画像及分析,通过用户画像可以了解到学生的相关属性、学习习惯等,也为后期的实验设计提供了依据;运用机器学习技术中的逻辑回归模型、增强决策树模型以及两种模型融合后的模型分别进行了学生成绩及格情况的预测,并根据预测结果分析了影响学生考试及格的因素。希望通过本文的研究,能够对大学英语教学和学习提供有价值的参考,帮助教师和学生有针对性的教学和学习。
[Abstract]:With the emergence of "Internet", all kinds of industries have undergone earth-shaking changes. "Internet education" has become a development trend, which has brought new opportunities and challenges to college English teaching and learning. Through the online English learning platform to carry on the teaching and the study, uses the data mining technology to carry on the analysis to the student study data and establishes the related model, explores the correlation between the English examination and each element. It is of great value to students' learning and teachers' teaching. The author has carried on the thorough research to this. This paper first describes the research status of data mining and college English teaching at home and abroad, and puts forward the research contents and methods of this topic. Secondly, the theoretical basis of this study is introduced, and the function, process and common machine learning model of data mining are introduced. Finally, the undergraduate students of Beijing University of posts and Telecommunications use the data generated by the college English teaching and evaluation system to construct and analyze the portraits of the students. Through the user portraits, we can understand the relevant attributes and learning habits of the students. It also provides the basis for the later experimental design. The logical regression model, enhanced decision tree model and the fusion model of the two models are used to predict the students' passing results, and the factors that affect the students' passing the examination are analyzed according to the predicted results. It is hoped that through this study, it can provide valuable reference for college English teaching and learning, and help teachers and students to have targeted teaching and learning.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：H319

【参考文献】