Support Vector Machine (SVM)

SVMs can optimally find categorization of linearly separable data and output whether a classification is true or not.

SVMs can be thought of as plotting all the data on a graph (2D) or n-dimensional data and figuring out how to separate data by drawing a line between the data points. It basically finds decision boundary that separates data points of different classes with the maximum margin. I.e. It finds the best line or curve that separates data points of different classes with the greatest possible distance.

Notes:

• SVM is often considered a Neural Network.
• SVM can operate in linear and non-linear data sets
• Support Vectors are the most useful data points because they are the most likely to be misclassified

SVM is used in:

• Classification
• Regression
• Outlier Detection

Types:

• SVM Classification
• SVM Regression