基于改进支持向量机的数学成绩预测模型  

Math scores prediction model based on improved support vector machine

作　　者：赵娜[1] 段志霞[1] 王慧[2] ZHAO Na;DUAN Zhi-xia;WANG Hui(Basic Department,Jiyuan Vocational and Technical College,Jiyuan 459000,China;Normal College,Nanyang Institute of Technology,Nanyang 473004,China)

机构地区：[1]济源职业技术学院基础部,河南济源459000 [2]南阳理工学院教师教育学院,河南南阳473004

出　　处：《白城师范学院学报》2023年第2期21-27,共7页Journal of Baicheng Normal University

摘　　要：文章提出一种改进蜂群算法(ABC)优化支持向量机(SVM)的数学成绩预测分析方法.首先,采用随机森林对输入变量进行特征提取.其次,通过基于二维均匀的种群初始化和欧氏距离的食物源更新对传统的人工蜂群算法进行改进,构建预测模型,进而提高SVM的预测性能.最后,通过使用UCI的数据集进行实验.实验结果表明,与其他优化算法对比,改进的SVM算法减少了数学成绩预测的计算量,缩短了预测时间,降低了预测错判率,提高了成绩预测的精确度,使数学成绩预测分析具有较好的准确率.This paper proposes an improved artificial bee colony algorithm(ABC)to optimize the predic-tion analysis method of the math scores of support vector machine(SVM).First of all,random forest is used to extract the features of input variables.Secondly,based on the two-dimensional homogeneous population initial-ization and food source update of Euclidean distance,the traditional artificial colony algorithm is improved,the prediction model is constructed,so as to improve the prediction performance of SVM.Finally,the experiment is carried out by using UCI data set.The experimental results show that compared with other optimization algo-rithms,this algorithm reduced the amount of calculation of prediction in math scores,shortens the time of prediction and reduces the rate of wrongful convictions in the prediction,improves the precision of the result predicted,and has a good accuracy for the prediction and analysis of math scores.

关 键 词：数学成绩 蜂群算法 支持向量机 准确率 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]