基于增强核极限学习机的专业选择智能系统  

An intelligent system based on enhanced kernel extreme learning machine for choosing the second major

作　　者：黄辉 冯西安[1] 魏燕[3] 许驰[3] 陈慧灵 HUANG Hui;FENG Xi-an;WEI Yan;XU Chi;CHEN Hui-ling(School of Marine Science and Technology,Northwestern Pol ytechnical University,Xilan 710072,China;College of Mathematics,Physics and Electronic Information Engineering,Wenzhou University,Wenzhou 325035,China;Wenzhou Vocational College of Science and Technology,Wenzhou 325006,China)

机构地区：[1]西北工业大学航海学院,西安710072 [2]温州大学数理与电子信息工程学院,浙江温州325035 [3]温州科技职业学院信息技术学院,浙江温州325006

出　　处：《吉林大学学报（工学版）》2018年第4期1224-1230,共7页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(61101155);浙江省自然科学基金项目(LY15F020033);温州市科技计划项目(2016R0002);浙江省教育厅科学研究基金项目(Y201533884);浙江省科技计划项目(2014C32031)

摘　　要：基于粒子群优化(PSO)的增强型核极限学习机(KELM)提出了一种有效的预测模型PSO-KELM来辅助第二专业选择。在PSO-KELM中,PSO策略确定KELM的最佳参数。PSO-KELM与其他两个竞争方法在学生专业选择数据上通过10折交叉验证方案进行比较,这两个方法分别是支持向量机和网格搜索技术优化的KELM。结果表明了本文预测模型在分类精度、受试者工作特征曲线面积(AUC)、灵敏度和特异性方面的优越性。This paper proposes an effective prediction model for choosing the second major based on the Particle Swarm Optimization（PSO）enhanced Kernel Extreme Learning Machine（KELM）,which is called PSO-KELM model.In this model,the PSO strategy is adopted to adaptively determine the optimal parameters in KELM.The PSO-KELM model is compared with other two competitive methods,including Support Vector Machine（SVM）and a KELM is optimized by grid search technique,on a major selection dataset via a 10-fold cross validation scheme.The results clearly confirm the superiority of the proposed PSO-KELM model in classification accuracy,area under the receiver operating characteristic curve（AUC）,sensitivity and specificity.

关 键 词：计算机应用 核极限学习机 粒子群优化 第二专业选择 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]