基于因子分析和迭代随机森林方法的学生成绩综合评价——以都匀市某高中为例

Comprehensive Evaluation of Students′scores Based on Factor Analysis and Iterative Random Forest Method--Taking a High School in Duyun City as an Example

新课标准明确提出教学评价体系中应建立多元化的目标,关注学生个性与潜能的发展。鉴于此,采用因子分析对学生的多学科成绩降维成少数几个典型因子,诸如理科因子水平和文科因子水平,依据学生在每个因子水平上的得分,可以更为全面的评测出学生在理科和文科成绩上的表现,以及是否存在偏科问题。在学生成绩等级评定的实证教学研究中,首次利用迭代随机森林方法,分别根据学生的各科成绩和反映学生成绩的因子水平,实现了自动对学生成绩进行等级评定,考虑了不同的迭代次数下,对迭代随机森林成绩评定精度作出了对比分析。

The new curriculum standards clearly put forward that the teaching evaluation system should establish diversified objectives,pay attention to the development of students′personality and potential.In view of this,factor analysis is adopted to reduce the multidisciplinary performance of students into a few typical factors,such as the factor level of science and the factor level of arts.Based on the score of each factor level of students,the performance of students in science and arts can be more comprehensively evaluated,as well as whether there is a problem of partial subjects.In student performance rating empirical teaching research,for the first time,the method of iterative random forest are used.Respectively according to the student′s grade and reflect the factors of the student′s grade level,it has realized automatic on student performance rating.With different iteration times taken into account,the evaluation accuracy of iterative random forest was compared and analyzed.