基于关联规则与聚类分析的课程评价技术

范圣法; 张先梅; 虞慧群

doi:10.14135/j.cnki.1006-3080.20201229001

基于关联规则与聚类分析的课程评价技术

doi: 10.14135/j.cnki.1006-3080.20201229001

1.
华东理工大学教务处
2.
计算机科学与工程系，上海 200237

基金项目: 国家自然科学基金(61702334, 61772200)；上海市自然科学基金(17ZR1406900, 17ZR1429700)；上海市高等教育学会规划课题(GJEL18135)

详细信息

作者简介:
范圣法（1984-）,男,安徽全椒人，博士生，助理研究员，主要研究方向为课程教学与评价。E-mail：shengfafan@ecust.edu.cn

通讯作者:
虞慧群，E-mail：yhq@ecust.edu.cn

中图分类号: TP391.4
计量
- 文章访问数: 177
- HTML全文浏览量: 121
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2020-12-29
- 网络出版日期: 2021-03-31

Curriculum Evaluation System Based on Association Rules and Cluster Analysis

1.
Office of Academic Affairs and
2.
Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai 200237, China

摘要

摘要: 高校在长期的教学活动中积累了大量的课程数据，如何利用数据资源分析课程教学状况，为提高课程教学质量提供决策支持，具有重要的研究价值。本文设计实现了一个基于关联规则与聚类分析的课程评价体系，对课程评价系统进行了功能需求分析，并对课程评价数据进行预处理。采用FP-growth算法对学生课程成绩数据进行关联规则分析，采用K-means++算法进行聚类分析，提高了课程数据分析的精度，实现了课程评价的自动化，提高了效率和评价的客观性。
- 课程评价 /
- 数据预处理 /
- FP-growth算法 /
- 关联规则 /
- 数据聚类
Abstract: The essential factor to consistently improve the teaching quality of college education relies on the quality of curriculum, which is the basis to realize the reform idea of all higher education. Colleges and universities have accumulated a large amount of curriculum data in long-term teaching activities. How to use data to evaluate the course and provide decision support for improving the quality of course teaching has become an important research field. This paper designs a curriculum evaluation system based on association rules and cluster analysis, then analyzes the functional requirements of the curriculum evaluation and preprocesses the course evaluation data. The FP-growth algorithm is used to analyze the association rules of the score of student course and K-means++ algorithm is used for cluster analysis, which improves the analysis accuracy of course data, realizes the automation of course evaluation, and improves the efficiency and objectivity of evaluation.
- curriculum evaluation /
- data preprocessing /
- FP-growth algorithm /
- association rules /
- data cluster

HTML全文

图 1 数据采集用例图

Figure 1. Use case of data acquisition

下载: 全尺寸图片幻灯片

图 2 报表6A-指标点相关考核项分解说明基本样式

Figure 2. Report 6A - basic styles of decomposition of relevant assessment items of indicator points

下载: 全尺寸图片幻灯片

图 3 数据分析及决策支持用例图

Figure 3. Use case of data analysis and decision-making support

下载: 全尺寸图片幻灯片

图 4 K-means++算法实例

Figure 4. Instances of K-means++ algorithm

下载: 全尺寸图片幻灯片

表 1 考试成绩离散化处理

Table 1. Discretization of examination score

Discretization judgment	Grade	Representation
Examination result>Average score	Excellent	1
Examination result≤Average score	Not excellent	0

下载: 导出CSV

表 2 课程关联规则

Table 2. Curriculum association rules

Antecedent	Consequent	Confidence level/%
Linear algebra、 Theory of probability、 Digital logic	Discrete mathematics	90.00
Assembly language、 Microcomputer principles	Operating system	96.29
College english、 Introduction to computer	Computer professional english	88.76
Assembly language、 Composition principle、 Computer architecture	Microcomputer principles	97.23
$ \vdots $	$\vdots $	$\vdots $

下载: 导出CSV

表 3 二元类型数据的列联表

Table 3. Contingency table of binary type data

	Object j
Object i		1	0	sum
	1	a	b	a+b
	0	c	d	c+d
	sum	a+c	b+d	a+b+c+d

下载: 导出CSV

表 4 第2个聚类中心的计算

Table 4. Calculation of the second clustering center

Sample number	D(x)²	P(x)	S
1	8	0.2	0.2
2	13	0.325	0.525
3	5	0.125	0.65
4	10	0.25	0.9
5	1	0.025	0.925
6	0	0	0.925
7	2	0.05	0.975
8	1	0.025	1

下载: 导出CSV

表 5 离散数学课程最终聚类中心

Table 5. Final clustering center ofdiscrete mathematics

Type	Standardized test scores	Standardized regular scores
First	0.9862	0.8824
Second	0.9598	0.7149
Third	0.9422	0.4379
Fourth	0.6000	0.5975

下载: 导出CSV

参考文献(12)

[1]	陈利民, 燕珍, 杜金涛, 等. 基于发展性评价的高校教师教学评价体系构建[J]. 微型电脑应用, 2019, 35(9): 125-127. doi: 10.3969/j.issn.1007-757X.2019.09.038
[2]	YE Q, JIAO M. The research of college computer courses teaching quality improvement[C]//2011 6th International Conference on Computer Science & Education. Singapore: IEEE, 2011: 572-576.
[3]	VARGAS H, HERADIO R, CHACON J, et al. Automated assessment and monitoring support for competency-based courses[J]. IEEE Access, 2019, 7: 41043-41051. doi: 10.1109/ACCESS.2019.2908160
[4]	乐惠骁, 范逸洲, 贾积有, 等. 优秀的慕课学习者如何学习——慕课学习行为模式挖掘[J]. 中国电化教育, 2019(2): 72-79. doi: 10.3969/j.issn.1006-9860.2019.02.012
[5]	赵辉. 数据挖掘技术在学生成绩分析中的研究及应用[D]. 辽宁大连: 大连海事大学, 2007.
[6]	朱跃. 数据挖掘在成绩分析中的应用性研究[D]. 贵阳: 贵州大学, 2009.
[7]	王婷婷. 数据仓库和数据挖掘在学生成绩分析中的应用研究[D]. 武汉: 武汉科技大学, 2007.
[8]	刘博鹏, 樊铁成, 杨红. 基于数据挖掘技术的学生成绩预警应用研究[J]. 四川大学学报(自然科学版), 2019, 56(2): 267-272.
[9]	ARTHUR D, VASSILVITSKII S. K-means++: The advantage of careful seeding[C]//18th ACM-SIAM Symposium on Discrete Algorithms. USA: ACM, 2007: 1027-1035.
[10]	ANWAR A, MAHMOOD A N, TARI Z. Identification of vulnerable lines in large power grid based on FP-growth algorithm[J]. Information Systems, 2015, 53: 201-212. doi: 10.1016/j.is.2014.12.001
[11]	YUAN J, TIAN Y. Practical privacy-preserving MapReduce based K-Means clustering over large-scale dataset[J]. IEEE Transactions on Cloud Computing, 2019, 7(2): 568-579. doi: 10.1109/TCC.2017.2656895
[12]	KARLEKAR A, SEAL A, KREJCAR O, et al. Fuzzy K-means using non-linear S-distance[J]. IEEE Access, 2019, 7: 55121-55131. doi: 10.1109/ACCESS.2019.2910195