基于偏最小二乘法的学生成绩预测研究
Research on student achievement prediction based on partial least square method
高艺璐 1刘鸿雁 1李劲松 2潘子尧3
作者信息
- 1. 渤海大学教育科学学院,辽宁锦州 121013
- 2. 渤海大学信息科学与技术学院,辽宁锦州 121013
- 3. 渤海大学数学科学学院,辽宁锦州 121013
- 折叠
摘要
成绩预测已经成为提高教学质量的重要手段之一.目前多数成绩预测方法都是单一的考虑课程成绩,而忽略了各门课程之间的相关性.考虑到课程之间的相互支撑关系,提出了基于偏最小二乘法的成绩预测方法,在传统成绩预测方法的基础上进一步充分考虑课程之间的相关性.为验证算法性能,以某校信息与计算科学专业的学生成绩为研究对象对算法进行了验证.实验结果表明,基于偏最小二乘法的成绩预测方法具有更低的预测误差,并且因其受训练样本数量影响较小,更易于泛化推广.
Abstract
Achievement prediction has become one of the important means to improve teaching quality.At present,most of the prediction methods only consider the course results and ignore the correlation between courses.Considering the mutual support between courses,a prediction method based on partial least square method is proposed,which further considers the correlation between courses on the basis of traditional predic-tion method.In order to verify the performance of the algorithm,the results of students majoring in informa-tion and computing science in a university are taken as the research object to verify the algorithm.The experi-ment results show that the performance prediction method based on partial least square method has lower pre-diction error,and it is easier to generalize because it is less affected by the number of training samples.
关键词
偏最小二乘法/成绩预测/教育数据挖掘/高等教育/显著性检验Key words
Partial Least Squares/achievement prediction/education data mining/tertiary education/sig-nificance test引用本文复制引用
基金项目
辽宁省教育科学研究项目(WJ2020004)
辽宁省教育科学研究项目(LJ2020003)
辽宁省教学改革项目(10167-636)
辽宁省教学改革项目(10167-649)
辽宁省教学改革项目(10167-800)
研究生教育教学改革项目(YJG20210002)
研究生教育教学改革项目(YJG20210005)
渤海大学本科教学改革研究项目(BHUXJGA2022030)
出版年
2024
NETL
NSTL
万方数据