Entity

Time filter

Source Type


Lin L.,Fujian Normal University | Yang S.,Fujian Jiangxia University | Zuo R.,Fujian Normal University
Proceedings of 2010 International Conference on Intelligent Control and Information Processing, ICICIP 2010 | Year: 2010

To improve the performance of secondary structure prediction, a multi-SVM ensemble was applied, bagging was used to resample the training dataset. The SVM ensemble was made of two-layer, one is composed by three SVM network decided by winner-take-all, the other is a ensemble network composed of five classifier decided by majority voting. Seven-fold cross-validation test on RS126 dataset indicated that the multi-SVM ensemble could achieve better performance on secondary structure prediction. © 2010 IEEE. Source


Bin S.,Fujian Jiangxia University
BioTechnology: An Indian Journal | Year: 2014

This paper is based on the investigation on the status quo of psychological contract among mittelstand enterprise staffs by introducing psychological contract to the field of mittelstand enterprises through literature review and compiling questionnaires. The findings show that to facilitate the management, the establishment of the corresponding psychological contract management measures should take into account of diverse needs of staffs from different age groups and with different length of service and different stages of their career development. On the basis of the above factors, the management measures can be explored from the facet of enterprise organization and individual employee. © Trade Science Inc. Source


Cai S.,Fujian Jiangxia University | Watanabe Y.,Kyushu University
Japan Journal of Industrial and Applied Mathematics | Year: 2015

We propose a computer-assisted method for excluding eigenvalues of an elliptic operator linearized at a solution of a nonlinear problem. The method works in both the one-dimensional and the two-dimensional case. We begin by finding an approximate solution to a nonlinear problem, and we then enclose the solution by using Nakao’s numerical verification method. Instead of considering directly the eigenvalues for the elliptic operator linearized at the verified solution, we linearize the operator at the approximate solution. We present a theorem that allows us to determine under which conditions and in which disks there will be no eigenvalues. Thus, if any of those disks are contained in the enclosed area, we can exclude those eigenvalues. Next, we construct various computable criteria that allow us to use a computer program to find these disks. Finally, we use our results to determine which eigenvalues to exclude for the operator linearized at the verified solution. We present some verified results. © 2015, The JJIAM Publishing Committee and Springer Japan. Source


Li X.,Fujian Jiangxia University | Yu Y.,Fujian Jiangxia University | Zheng Z.,Fujian Normal University
Ceramics International | Year: 2015

The upconversion emission of Er3+:PLZT transparent ceramic was investigated at different temperatures from 10 to 300K. Emission bands around 540, 564 and 682nm corresponding to transitions from 2H11/2, 4S3/2, and 4F9/2 to 4I15/2 were observed. As the measurement temperature was increased, the intensities of the 564 and 682nm peaks decreased, while that of the 540nm peak increased. Based on the rate equation theory, a model for the dynamics of upconversion emission in Er3+:PLZT ceramic was proposed. Impressively, the calculations agree well with the experimental data in the whole temperature range. Additionally, the temperature dependent upconversion emission behavior in Er3+:PLZT ceramic was discussed in terms of the Boltzmann thermal distribution and nonradiative relaxation. © 2015 Elsevier Ltd and Techna Group S.r.l. Source


Lin L.,Fujian Normal University | Zuo R.,Fujian Normal University | Yang S.,Fujian Jiangxia University | Zhang Z.,Fujian Jiangxia University
ICICIP 2012 - 2012 3rd International Conference on Intelligent Control and Information Processing | Year: 2012

Due to the expansion of high-speed Internet access, the need for secure and reliable networks has become more critical. The sophistication of network attacks, as well as their severity, has also increased recently. In this paper, a new intelligent intrusion detection system has been proposed using SVM ensemble. The ensemble was made of two-layer, one is composed by five SVM network decided by winner-take-all, the other is a ensemble network composed of five classifier decided by majority voting. The KDD99 data sets was used to test which achieve a better performance. © 2012 IEEE. Source

Discover hidden collaborations