TY - GEN
T1 - Sample selection based on k-l divergence for effectively training SVM
AU - ZHAI, Junhai
AU - LI, Ta
AU - LI, Chang
AU - WANG, Xizhao
N1 - This research is supported by the national natural science foundation of China (61170040), by the natural science foundation of Hebei Province (F2013201110, F2013201220), by the Key Scientific Research Foundation of Education Department of Hebei Province (ZD2010139), by the natural science foundation of Hebei University (2011-228), and by the research projects on reform of education and teaching of Hebei University (JX07-Y-27).
PY - 2013
Y1 - 2013
N2 - The computational time and space complexity of support vector machine (SVM) are O(n3) and O(n2) respectively, where n is the number of training samples. It is inefficient or impracticable to train an SVM on relatively large datasets. Actually, the removal of training samples that are not support vector (SVs) has no effect on constructing the optimal hyperplane. Based on this idea, this paper proposed a sample selection method which can efficiently choose the candidate SVs from original datasets. The selected samples are used to train SVM. The experimental results show that the proposed method is effective and efficient; it can efficiently reduce the computational complexity both of time and space especially on relatively large datasets.
AB - The computational time and space complexity of support vector machine (SVM) are O(n3) and O(n2) respectively, where n is the number of training samples. It is inefficient or impracticable to train an SVM on relatively large datasets. Actually, the removal of training samples that are not support vector (SVs) has no effect on constructing the optimal hyperplane. Based on this idea, this paper proposed a sample selection method which can efficiently choose the candidate SVs from original datasets. The selected samples are used to train SVM. The experimental results show that the proposed method is effective and efficient; it can efficiently reduce the computational complexity both of time and space especially on relatively large datasets.
KW - K-l divergence
KW - Pnn
KW - Samples selection
KW - SVM
UR - http://www.scopus.com/inward/record.url?scp=84893613151&partnerID=8YFLogxK
U2 - 10.1109/SMC.2013.823
DO - 10.1109/SMC.2013.823
M3 - Conference paper (refereed)
AN - SCOPUS:84893613151
SN - 9780769551548
T3 - IEEE International Conference on Systems, Man and Cybernetics
SP - 4837
EP - 4842
BT - Proceedings : 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013
PB - IEEE
T2 - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013
Y2 - 13 October 2013 through 16 October 2013
ER -