Please be informed that SVMs are different from other classification algorithms because of the way they select the decision boundary that maximizes the length from the nearest data points of all the classes. The decision boundary is being generated by SVMs that is known as the maximum margin classifier or it is also known as the maximum margin hyperplane.
SVM uses a subset of training data points in the decision function which is known as support vectors which makes it memory efficient.
There are several kernel functions that can be defined for the decision function. In that case, you can use common kernels.
For more information regarding the same, do refer to the Machine Learning Certification course.