稀疏最小二乘支持向量机及其应用

Based on feature vector selection (FVS) method, a new model of sparse least squares support vector machine (SLS-SVM) is proposed to solve the sparseness problem of least squares support vector machine (LS-SVM). Firstly, a subset of feature vectors is defined in feature space to reconstruct all the training samples linearly. Then, the sparse feature vectors are used as support vectors, and the SLS-SVM model is obtained. Finally, simulation experiment of pantograph-catenary system is carried out. The simulation results show that compared with basic LS-SVM and traditional SVM, SLS-SVM not only has good forecast precision, but also achieves highly sparse support vectors, for which the prediction speed is accelerated.