Open Science Research Excellence
%0 Journal Article
%A Tatjana Eitrich and  Bruno Lang
%D 2007 
%J  International Journal of Computer and Information Engineering
%B World Academy of Science, Engineering and Technology
%I International Science Index 11, 2007
%T Efficient Implementation of Serial and Parallel Support Vector Machine Training with a Multi-Parameter Kernel for Large-Scale Data Mining
%U http://waset.org/publications/13987
%V 11
%X This work deals with aspects of support vector learning for large-scale data mining tasks. Based on a decomposition algorithm that can be run in serial and parallel mode we introduce a data transformation that allows for the usage of an expensive generalized kernel without additional costs. In order to speed up the decomposition algorithm we analyze the problem of working set selection for large data sets and analyze the influence of the working set sizes onto the scalability of the parallel decomposition scheme. Our modifications and settings lead to improvement of support vector learning performance and thus allow using extensive parameter search methods to optimize classification accuracy.

%P 3598 - 3603