Support Vector Machines (page 10 of 10)

A few advantages of this algorithm are: 1) It performs very well on a range of datasets, 2) It is versatile, which means that different kernel functions may be specified (hyper-parameter), or custom kernels may be defined for specific datatypes, & 3) It works well for both high and low dimensional data (i.e., number of features).

A few disadvantages of this algorithm are: 1) The training (run) time and memory usage decrease as the size of the training dataset increases, 2) Needs careful normalization of input data and hyper-parameter tuning, 3) Does not provide a direct probability estimator, & 4) Difficult to interpret why a prediction was made.

InClass Example: Predicting Sales Call Outcome