Title :
The combination evaluation method based on DEA and SVM
Author :
Niu, Dong-xiao ; Wang, Xiu-mei ; Gao, Chong
Author_Institution :
Sch. of Bus. & Adm., North China Electr. Power Univ., Baoding
Abstract :
SVM has the convenient superiority in the classification. But the insufficiency is required the classed samples ahead of time. This affects its widespread application. The paper tries to take full of DEA methodpsilas objective advantages and SVM modelpsilas machine learning advantages. Using DEA method classes the samples data, so as to the training samples have strong reliability. Then, multi-layer SVM is used to evaluate teaching efficiency. In order to verify the effectiveness of the method, 40 faculties are used, which include 34 samples and 6 for experiment. The experimental results show that this method has strong objectivity and impartiality, the combination evaluation method is simple and easy to popularize.
Keywords :
data envelopment analysis; support vector machines; data envelope analysis; machine learning; multilayer SVM; samples data; support vector machine; Cybernetics; Data analysis; Data envelopment analysis; Decision making; Education; Electronic mail; Learning systems; Machine learning; Support vector machine classification; Support vector machines; Data envelope analysis (DEA); Evaluation; Support vector machines (SVM);
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620522
