What are Support Vector Machines?
A support vector machine is a collection of supervised learning algorithms that use hyperplane graphing to analyze new, unlabeled data. These machines are mostly employed for classification problems, but can also be used for regression modeling.
How are Support Vector Machines Used?
In the most common application, incoming data is categorized into one of two classes. For SVM models, each data point is interpreted as a p-dimensional vector, which the machine attempts to create a linear classifier by fitting the data point inside a hyperplane (p-1 dimension).
So for classification, every input is turned into a point in n-dimensional space (n being the number of features), and the value of each feature defined as the value of a unique coordinate on the hyperplane. Classification is determined by finding the hyperplane that most clearly separates the two classes. For example:
