Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Education and Culture Executive Agency (EACEA). Neither the European Union nor EACEA can be held responsible for them.


Support Vector Machine (SVM)

Users: 1 - Average Rating: 5.00

Support vector machines (SVMs) belong to the family of the kernel methods.
The main features of SVM are the following:
• It is a supervised learning method which can be used for classification, regression, and outliers’ detection.
• It exploits the “kernel trick” for distances in order to extend linear techniques to the solution of non-linear problems.
• It is effective in high dimensional spaces and uses a subset of training points in the decision function (called support vectors), so it is also memory efficient.
• As kernel methods, it selects the model which minimizes the trade-off between the performance over the data and the complexity of the solution.
• The main hyperparameters of the SVM algorithm are the kernel’s type (which is usually the Gaussian one for kernel SVM), kernel’s hyper-parameter γ, and the regularization’s parameter usually indicated as C or α. The regularization parameter controls the complexity of the solution, while the kernel’s hyperparameter controls the non-linearity of the solution.
• SVM does not handle categorical features directly (consequently encoding is needed).
• SVM suffers from numerical issues and consequently input data must be scaled.

Scientific Area:


Target Group:

Cite as:
Christianini, N. and Shawe-Taylor, J. C., An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods, Cambridge University Press, 2000.

Author of the review:
Giulia Cademartori
University of Genoa


You have to login to leave a comment. If you are not registered click here

Maria de Fátima Pacheco

Nice and useful implementation of SVM!