What is a support vector?
A support vector is a collection of data points that serve as boundaries within a dataset on the separating hyperplane. They form an area that include other points that are similarly related in distance to each other.
What is the Kernel trick?
The kernel trick is a streamlined equation that allows us to perform operations on n-dimensional data. It uses dot product math to compute the shortcut within our data, arriving at the same conclusion.
What is the hyper plane and how is it utilized in SVMs?
The hyperplane is the 3-D dividing line representing the decision points on how classes are divided. Support vector machines make the decision once the support vector boundaries are outlined by drawing a “line/plane” where that decision has been made from grouped classes. The hyperplane/boundary is determined where classes/datapoints are no longer likely to be similar.