Suykens J.A.K., Van Gestel T., De Brabanter J., De Moor B., Vandewalle J. Least Squares Support Vector Machines

World Scientific Publishing Co. Pte. Ltd. 2005. - 310 p.
Driven by the dream to make the approach as simple as possible (but not simpler) led us to the formulation of least squares support vector machine classifiers, as a first contribution in this area. Many tests and comparisons showed great performance of LS-SVMs on several benchmark data set problems and were very encouraging for further research in this promising direction. At the ESAT-SISTA research division of the Electrical Engineering department of the Katholieke Universiteit Leuven a lot of expertise in the area of mathematical engineering, including neural networks, was available which largely motivated the study of least squares support vector machines from this perspective. Conceptually, the additional explicit primal-dual interpretations from the viewpoint of optimization theory turned out to be essential for further developments of least squares support vector machines. In the neural networks area the emphasis has always been on universal models with applications within a very broad context such as function estimation, recurrent modeling, classification, control, unsupervised learning, on-line learning and many more.