大数据下最优教学方式选取模型设计  

Design of selection model for optimal teaching mode under big data

作　　者：许健松[1] 游晓东[2] XU Jian-song;YOU Xiao-dong(Office of Academic Affairs,Fujian Agriculture and Forestry University,Fuzhou 350002,China;School of Management,Fujian Agriculture and Forestry University,Fuzhou 350002,China)

机构地区：[1]福建农林大学教务处,福州350002 [2]福建农林大学管理学院,福州350002

出　　处：《沈阳工业大学学报》2018年第6期664-669,共6页Journal of Shenyang University of Technology

基　　金：福建省教育科学"十三五"规划项目(FJJKCGZ17-101)

摘　　要：为了获取关联性较强、数据融合精度较高的最优教学模式,提出依据数据估计量变化的最优教学方式选取模型设计方法.对不同服务器大数据信息进行分段,利用前半段数据进行曲线拟合,后半段数据进行直线拟合.预测高噪声干扰数据,分析干扰数据点滤除前后统计量的变化,构建由数据源整合、数据拟合、滤除干扰三方面相结合的最优教学方式选取模型并进行相关性计算.利用调整系数进一步提高了模型精度.结果表明,所设计模型的表达能力较强,复杂度不高,可精准、快速地实现最优教学方式的选取.In order to obtain the optimal teaching mode with strong relevance and high data fusion accuracy,a design method of selection model for optimal teaching mode based on the change of data estimation was proposed.With the segmentation of large data information from different servers,the curve fitting was carried out with the first half data,and the line fitting was performed with the second half data.The interference data with high noise were predicted,and the change of statistics before and after filtering the interference data points was analyzed.In addition,the selection model for the optimal teaching mode in combination with the data source integration,data fitting and interference filtering was established,the correlation calculation was performed,and the model accuracy was further enhanced with the adjustment coefficients.The results show that the presentation ability of designed model is strong,and the complexity is not high,which can accurately and quickly achieve the selection of optimal teaching mode.

关 键 词：大数据 数据拟合 滤除干扰 模型设计 数据挖掘 分段处理 特征选取 教学方式 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]