面向学生成绩预测的组合优化算法  

Combined optimization algorithm for student achievement prediction

作　　者：党佳俊 张宏烈[1] 慕钢 李诚[1] 张晓琳 DANG Jiajun;ZHANG Honglie;MU Gang;LI Cheng;ZHANG Xiaolin(School of Computer and Control Engineering,Qiqihar University,Qiqihar 161006,China;Department of Information Technology,Teacher Training School in Jianhua District of Qiqihar,Qiqihar 161006,China)

机构地区：[1]齐齐哈尔大学计算机与控制工程学院,黑龙江齐齐哈尔161006 [2]齐齐哈尔市建华区教师进修学校信息技术部,黑龙江齐齐哈尔161006

出　　处：《高师理科学刊》2022年第5期40-46,共7页Journal of Science of Teachers＇College and University

基　　金：黑龙江省教育厅基本业务专项理工面上项目(135509118)。

摘　　要：利用机器学习算法分析和预测学生成绩是大数据技术应用之一.将启发式算法与梯度提升算法相结合,提出组合优化算法预测模型.首先,通过采用动态对立学习增加种群初始化的多样性,引入非线性收敛因子和自适应权重等方法,得到增强鲸鱼算法,改进原来的全局搜索和局部开发能力.其次,基于XGboost模型加以增强鲸鱼算法的迭代,动态优化XGboost的超参数,提出组合算法预测模型.准确率ACC作为模型的评价标准,以学生数据集为研究对象,以学生成绩预测为目标,选用5种算法进行对比实验.实验结果表明,组合算法的预测准确度相对较高.It is one of applications of big data technology to analyze and predict student achievement by using machine learning algorithms.The prediction model of a combined optimization algorithm is proposed,in which the heuristic algorithm and the gradient boosting algorithm are combined.First of all,the dynamic oppositional learning is adopted to increase the diversity of population initialization,and nonlinear convergence factors and adaptive weights are introduced.Thereby,the enhanced whale algorithm is obtained,and the original global search and the local development capabilities are improved.Then,based on the XGboost model,the iteration of the whale algorithm is enhanced,the hyper-parameters of XGboost are dynamically optimized,and a combined algorithm prediction model is proposed.Accuracy ACC is used as the evaluation standard of the model,taking the student data set as the research object and the student achievement prediction as the goal.Five algorithms are selected for the comparative experiments.The experimental results verify that the prediction accuracy of the combined algorithm proposed is relatively higher.

关 键 词：组合优化算法 增强鲸鱼算法 XGboost算法 学生成绩预测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]