Table of Contents#### Download Safari Books Online apps: Apple iOS | Android | BlackBerry

# Support-Vector Machines

Entire Site

Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

Consider again the challenge of finding a straight line that divides two classes. Figure 9-9 illustrates an example. The averages for each class are shown in the figure, along with the dividing line that they imply.

Notice that the dividing line calculated by using the averages misclassifies two of the points because they are much closer to the line than the majority of the data. The problem is that since most of the data is far away from the line, it’s not relevant for including in the dividing line.

Support-vector machines are a well known set of methods for
creating classifiers that solve this problem. They do this by trying to
find the line that is as far away as possible from each of the classes.
This line is called the *maximum-margin hyperplane*,
and is shown in Figure 9-10.