Title :
Design of support vector machine by adaptive aggregation
Author :
Chacon, Oscar ; Litvintchev, Igor ; Alvarez, Ada ; Vazquez, Ernesto
Author_Institution :
Graduate Program of Syst. Eng., Univ. Autonoma de Nuevo Leon, Mexico
Abstract :
This article provides a new algorithm to solve the design of classification machine, for linearly separable sets, based in support vectors. For large scale binary classification, an adaptive aggregation (AAM) procedure is executed so that the size of possible support vectors decrease, in each iteration, until convergence to maximum separation margin is achieved.
Keywords :
convergence of numerical methods; iterative methods; learning (artificial intelligence); optimisation; pattern classification; support vector machines; adaptive aggregation; classification machine design; convergence; iterative methods; large scale binary classification; maximum separation margin; support vector machine design; support vectors; Active appearance model; Algorithm design and analysis; Convergence; Large-scale systems; Pattern recognition; Statistical learning; Statistics; Support vector machine classification; Support vector machines; Systems engineering and theory;
Conference_Titel :
Neural Networks, 2003. Proceedings of the International Joint Conference on
Print_ISBN :
0-7803-7898-9
DOI :
10.1109/IJCNN.2003.1223729
