论文信息 - A New Formulation for Classification by Ellipsoids

A New Formulation for Classification by Ellipsoids

We propose a new formulation for the optimal separation problems. This robust formulation is based on finding the minimum volume ellipsoid covering the points belong to the class. Idea is to separate by ellipsoids in the input space without mapping data to a high dimensional feature space unlike Support Vector Machines. Thus the distance order in the input space is preserved. Hopfield Neural Network is described for solving the optimization problem. The benchmark Iris data is given to evaluate the formulation.

[1] Zhihua Zhang,et al. Learning Metrics via Discriminant Kernels and Multidimensional Scaling: Toward Expected Euclidean Representation , 2003, ICML.

[2] WU Vincent. SVM and CO-TRAINING , 2002 .

[3] Stephen P. Boyd,et al. Applications of semidefinite programming , 1999 .

[4] Malte Kuss. Nonlinear Multivariate Analysis with Geodesic Kernels , 2002 .

[5] Aníbal R. Figueiras-Vidal,et al. Sample selection via clustering to construct support vector-like classifiers , 1999, IEEE Trans. Neural Networks.

[6] Annabella Astorino,et al. Ellipsoidal separation for classification problems , 2005, Optim. Methods Softw..

[7] E. Barnes. An algorithm for separating patterns by ellipsoids , 1982 .

[8] Robert P. W. Duin,et al. Support vector domain description , 1999, Pattern Recognit. Lett..

[9] Vladimir Vapnik,et al. Statistical learning theory , 1998 .

[10] Nello Cristianini,et al. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[11] F. Glineur. Pattern separation via ellipsoids and conic programming , 1998 .

[12] Nello Cristianini,et al. An introduction to Support Vector Machines , 2000 .