摘要
学生成绩预测是改进教学、提高教学质量的一种重要辅助方法。目前成绩预测研究多数是基于全部训练样本构建预测模型,训练样本没有针对性,未能考虑类内样本的相似性和类间样本的差异性对预测模型的构建和预测效果的影响。因此,为了充分利用类内样本的相似性特性,提出一种基于K-Means的学生成绩预测方法。以某学校汉语言文学和计算机应用两门专业学生的课程成绩为对象,经过大量的实验证明,该方法构建的成绩预测模型具有更高的预测精度,进而可以为改进教学提供更为准确的参考信息,辅助提升学校的教学质量。
Student score prediction is an important auxiliary method to improve teaching and teaching quality.At present,most of the score prediction studies are based on all training samples to build prediction models.The training samples are not targeted,and the similarity and difference of samples are not considered.Therefore,in order to make full use of the similarity characteristics of samples,a student score prediction method based on K-Means is proposed.The course scores of Chinese language literature and computer application majors in a school are taken as the object,and a large number of experiments proves that the score prediction model constructed by this method has higher prediction accuracy,which can provide more accurate reference information for improving teaching and the teaching quality of the school.
作者
王冠帮
刘鸿雁
李劲松
刘晓雲
WANG Guan-bang;LIU Hong-yan;LI Jin-song;LIU Xiao-yun(School of Education Science,Bohai University,Jinzhou 121000,Liaoning Province,China;School of Information Science and Technology,Bohai University,Jinzhou 121000,Liaoning Province,China)
出处
《信息技术》
2023年第2期1-6,共6页
Information Technology
基金
辽宁省自然科学基金项目(2019-ZD-0503)
辽宁省教育厅科学研究项目(WJ2020004,LJ2020003)。