SVM-KNN算法在机电职业学习者属性特征智能预测中的应用研究  

Application of SVM-KNN algorithm in intelligent prediction of attribute characteristics of electromechanical vocational learners

作　　者：牛全峰[1] NIU Quanfeng(ENSHI Polytechnic,Enshi Hubei 445000,China)

机构地区：[1]恩施职业技术学院,湖北恩施445000

出　　处：《自动化与仪器仪表》2023年第3期76-79,共4页Automation & Instrumentation

基　　金：2020年湖北省高等学校省级教学研究项目《“三全育人”视域下高职院校校企融合推进人才培养路径研究》(2020835)。

摘　　要：现代社会环境下机电职业学习者特征与其他学科学习者特征存在着显著差别,为了对机电职业学习者属性特征进行智能预测,提出一种结合SVM与优化KNN的算法模型。该模型首先对传统KNN模型进行加权来应对处理数据时的不均衡问题;在此之后结合SVM算法和加权KNN算法各自的优点对机电职业学习者样本进行分类,即距离超平面分类较远距离的学习者样本选择SVM算法,较近距离的则采用加权KNN算法。实验结果表明,融合各自优点的SVM-KNN算法具有更高的数据分类准确率,对机电职业学习者的属性特征分类和预测具有较好的适用性。In the modern social environment, the characteristics of electromechanical vocational learners are significantly different from those of learners in other disciplines. In order to intelligently predict the attribute characteristics of electromechanical vocational learners, this paper proposes an algorithm model combining SVM and optimized KNN. Firstly, the traditional KNN model is weighted to deal with the imbalance of data processing;After that, combine the respective advantages of SVM algorithm and weighted KNN algorithm to classify the electromechanical vocational learner samples, that is, the learner samples that are far from the hyperplane classification select SVM algorithm, and those that are close to the hyperplane classification use weighted KNN algorithm. The experimental results show that the SVM-KNN algorithm combined with their respective advantages has a higher accuracy rate of data classification, and has a good applicability to the attribute feature classification and prediction of electromechanical vocational learners.

关 键 词：机电职业 SVM-KNN算法 学习者特征 智能预测 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]