Directly Optimised Support Vector Machines for Classification and Regression
- Publisher: Department of Automatic Control and Systems Engineering
A new method of implementing Support Vector learning algorithms for classification and regression is presented which deals with problems of over-defined solutions and excessive complexity. Classification problems are solved with the minimum number of support vectors, irrespective of over-lapping training data. Support vector regression can be solved as a sparse solution, without requiring an e-insensitive zone. The optimisation method is generalised to include control of sparsity for both support vector classification and regression.
views in local repository
downloads in local repository
The information is available from the following content providers: