Support vector machine (machine learning)

Last revised by Candace Makeda Moore on 26 Aug 2020

The support vector machine (SVM) is a supervised learning algorithm used to separate groups of data with a margin or plane which is made as well as possible to ensure it is more likely to generalize well to examples it has never seen before. In the case of a two feature data set a margin or line can be used to separate different samples in the training data. Samples in the training data near the line, plane or hyperplane are termed support vectors.

There are different margins, planes or hyperplanes that can be used to achieve separation of groups of data. The SVM tries to maximize the orthogonal distance from the planes to the support vectors in each classified group. The SVM model is able to function i.e. find a useful margin or hyperplane best when there are values in the features of two groups that tend to group around different values e.g. predicting values related to a tumor grade or classifying different tissues with different attenuation and textures shown. Support vector machines have been applied successfully to both some image segmentation and some image classification problems. 

ADVERTISEMENT: Supporters see fewer/no ads

Updating… Please wait.

 Unable to process the form. Check for errors and try again.

 Thank you for updating your details.