基于机器学习的高校学生成绩预测
Grade prediction of university student based on machine learning
李凯伟1
作者信息
- 1. 山西科技学院大数据与计算机科学学院,山西 晋城 048000
- 折叠
摘要
以某校本科生的历史成绩数据、一卡通消费数据、校园网日志数据和图书馆刷卡记录数据为基础,提出一种利用学生行为数据来预测学生成绩的方法.选择五种常用于教育数据挖掘的预测方法(逻辑回归算法、支持向量机算法、决策树算法、K近邻算法和朴素贝叶斯算法),通过Stacking集成进行模型优化,实验结果表明,相较于单独利用成绩和单独利用分类模型预测成绩,该方法准确率更高,该研究对于辅助教学管理,促进智慧校园建设有一定意义.
Abstract
Based on the historical performance data,one card consumption data,campus network log data,and library card swiping record data of undergraduate students in a certain university,a method for predicting student grades using student behavior data is proposed.Five commonly used prediction methods(logistic regression algorithm,support vector machine algorithm,decision tree algorithm,K-nearest neighbor algorithm,and naive Bayesian algorithm)for educational data mining are selected for model optimization through Stacking integration.Experimental results show that this method has higher accuracy compared to predicting grades using grades alone and classification models alone.This study has certain significance for assisting teaching management and promoting the construction of smart campuses.
关键词
教育数据挖掘/成绩预测/组合优化Key words
education data mining/grade prediction/combination optimization引用本文复制引用
基金项目
山西科技学院校内项目基于机器学习的成绩预警(XKY001)
出版年
2023
NETL
NSTL
维普
万方数据