Papers by Keyword: Support Vector

Abstract: With the development of electric power system, the people pay more attention to demand-side management. Valid for load classification is an important prerequisite for improving demand side management level. Based on common indicators and load clustering algorithms, new SVC algorithm, cluster validity analysis and similarity measurement of the impact of the judgment are proposed based on load clustering method, and finally the effectiveness of the method is demonstrated by an example.

Abstract: Intrusion detection algorithm based on support vector machine with pre-extracting support vector is proposed which combines the center distance ratio and classification algorithm. Given proper thresholds, we can use the support vector as a substitute for the training examples. Then the scale of dataset is decreased and the performance of support vector machine is improved in the detection rate and the training time. The experiment result has shown that the intrusion detection system(IDS) based on support vector machine with pre-extracting support needs less training time under the same detection performance condition.

Abstract: Model reference adaptive control (MRAC) is widely used in linear system control areas, and Neural Networks (NN) is often used to extend MRAC to nonlinear areas. However, this kind of solution inherits some drawbacks of NN, including slow learning speed, weak generalization ability, local minima tendency, etc. Given these drawbacks, this paper attempts to use support vector regression (SVR) as a substitute of NN. In this approach, SVR is employed to compensate the nonlinear part of the plant. A stable controller-parameter adjustment mechanism is constructed by using the practical stability theory. Simulation results show that the proposed approach could reach desired performance.

Abstract: In order to improve the recognition rate of the electronic nose system for small samples, an electronic nose pattern recognition algorithm based on support vector machine (SVM) is proposed in this paper. Identification experiments for three kinds of wine with similar odor were carried out. The sensor arrays are optimized by means of principal component analysis (PCA) method first. Then, make comparing experiment using different algorithms for different number of training samples of wine. The related results show that PCA-SVM based pattern recognition algorithms has high recognition accuracy, stronger classification capability, and has potential advantages in small sample classification and recognition experiments.

Abstract: Intrusion detection algorithm based on support vector machine with pre-extracting support vector is proposed which combines the center distance ratio and classification algorithm. Given proper thresholds, we can use the support vector as a substitute for the training examples. Then the scale of dataset is decreased and the performance of support vector machine is improved in the detection rate and the training time. The experiment result has shown that the IDS based on support vector machine with pre-extracting support needs less training time under the same detection performance condition.

Abstract: In order to figure out the deficiency of the SVM on extensive sample, nature of SV is studied in this paper. An improved incremental training algorithm is put forward based on dimensional of samples. A chosen gene which got by density and distance criterion is used in this method. In this method the number of training samples is decreased and the space information is keeped. So, the training speed is improved while the precision is not reduced. And the simulation proved the efficiency of this method.

Abstract: Improved learning algorithm for branch and bound for semi-supervised support vector machines is proposed, according to the greater difference in the optimal solution in different semi-supervised support vector machines for the same data set caused by the local optimization. The lower bound of node in IBBS3VM algorithm is re-defined, which will be pseudo-dual function value as the lower bound of node to avoid the large amount of calculation of 0-1 quadratic programming, reducing the lower bound of each node calculate the time complexity; at the same time, in determining the branch nodes, only based on the credibility of the unlabeled samples without the need to repeatedly carry out the training of support vector machines to enhance the training speed of the algorithm. Simulation analysis shows that IBBS3VM presented in this paper has faster training speed than BBS3VM algorithms, higher precision and stronger robustness than the other semi-supervised support vector machines.