|Boundary Support Vectors|
A support vector machine identifies a hyperplane (or a line in 2D space) which linearly separates the observations that belong to different classes. Those observations that are closest to the hyperplane are called the support vectors. The concept of a margin is used to maximise the distance between the support vectors.
Data is not usually linearly separable. Thus some kind of mapping of the data to higher dimensions is performed to find a space where it is possible to separate the classes with a hyperplane. A so-called kernel function is used to do this. Common kernel functions include radial basis function, linear, polynomial, and sigmoid.
See Section 18.17 for the one-class support vector machine useful for outlier detection.
% Free and Open Source Data Science Software Survival Guide % % Time-stamp: <Monday 2021-11-22 05:50:14 AEDT Graham Williams> % % Author: Graham.Williams@togaware.com % License: Creative Commons Attribution License 4.0 International % License: https://creativecommons.org/licenses/by/4.0/ % Copyright: (c) 1995-2021 Graham.Williams@togaware.com % % You are free to Share (copy and redistribute the material in any % medium or format) and Adapt (remix, transform, and build upon the % material for any purpose, even commercially), under specific terms % of Attribution (you must give appropriate credit, provide a link to % the license, and indicate if changes were made and you may do so in % any reasonable manner, but not in any way that suggests the licensor % endorses you or your use) and with No Additional Restrictions (you % may not apply legal terms or technological measures that legally % restrict others from doing anything the license permits).
Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0