Xie X.-L.,China and Guangxi Key Laboratory of Spatial Information and Geomatics |
Xie X.-L.,Guilin University of Technology |
Liao Z.-Y.,Guilin University of Technology |
Cai G.-Y.,Guilin University of Electronic Technology
Lecture Notes in Electrical Engineering | Year: 2014
The values of the support vector machines (SVM) model parameters and the kernel function parameters decisively affect the classification accuracy; however, mostly, the values of those parameters are random values or the values of experience which results in the low classification accuracy in order to improve the accuracy and efficiency of data classification and this paper uses the elastic cloud-computing cluster to provide computing power for faster calculation speed and also introduces particle swarm optimization (PSO) algorithm based on the optimization theory to optimize the parameters of the classifier in the SVM classification algorithm making the accuracy of the classifier as the fitness function of PSO algorithm to find the global optimum parameter values of SVM model and kernel function. The experimental results show that in the open source cloud-computing platform hardtop data classification accuracy has significantly improved. © Springer-Verlag Berlin Heidelberg 2014.
Xie X.,Guilin University of Technology |
Cai G.,China and Guangxi Key Laboratory of Spatial Information and Geomatics |
Cai G.,Guilin University of Electronic Technology
Applied Mechanics and Materials | Year: 2013
The paper proposes two application modes which are "Services request model" and "services provided - Request mode" in the pervasive environment. The detail architecture of manufacturing Grid based on "services provided - Request mode" is presented and analyzed. © (2013) Trans Tech Publications, Switzerland.