基于组合核函数的高校经济困难生分类

Classification of College Students with Financial Difficulties Based on Combination Kernel Function

为进一步提高高校资助工作的精准度,构建基于组合核函数的支持向量机(SVM)高校经济困难生分类模型。根据在校生的消费数据、人员信息及历史资助信息抽取样本特征,利用径向基(RBF)核函数的局部拟合能力及多项式核函数的泛化能力,构建基于RBF核函数及多项式核函数的组合核函数SVM分类模型;采用多重网格搜索法训练模型获取最优核参数和组合核函数的权系数,并对高校经济困难生进行分类预测。实验结果表明:采用构建的模型可对高校经济困难生进行分类预测,与单核核函数SVM、逻辑回归模型、最近邻算法(KNN)相比,其分类准确率显著提升;使用融合特征可增加不同类别样本数据的差异性,有助于提高分类准确率。

In order to further improve the accuracy of college financial aid,a support vector machine(SVM)classification model for college students with financial difficulties based on combined kernel function was constructed.According to the consumption data,personnel information and historical funding information of the students in school,sample features are extracted.Using the local fitting ability of radial basis function(RBF)kernel and the generalization ability of polynomial kernel,a SVM classification model based on RBF kernel and polynomial kernel was constructed;The multi-grid search method was used to train the model to obtain the optimal kernel parameters and the weight coefficients of the combined kernel function,and to classify and predict the students with financial difficulties.The experimental results show that the constructed model can be used to predict the financial difficulties of college students,and compared with single kernel SVM,logistic regression model and nearest neighbor algorithm(KNN),the accuracy of classification is significantly improved;Using fusion features can increase the difference of different types of sample data,and help to improve the accuracy of classification.