orithm and sequential minimal optimization.РWe propose a general framework, named MCPVC for recognizing radar high resolution rage profiles. pare support vector machines and the classifiers developed in the dissertation paper on the measured airplane data. The results show that MCPVC is superior to the two known time-shift invariant feature extraction method; FSALS-SVM developed in chapter 6 obtains the better generalization performance than support vector machines.РKey words: Kernel Methods, Support Vector Machines, Support Vector Classification, Support Vector Regression, Least Squares Support Vector Machine, Kernel Fisher Discriminant Analysis, LASSO, Fast Algorithms, Sequential Minimal Optimization, Greedy Stagewise Algorithm, Sparse Approximation, Sequential Sparse Greedy Optimization
全文预览
