You are on page 1of 1

Support vector machine:

A support vector machine is a classifier which classify data items based on the hyperplane that
was made by using some labelled data. This hyperplane will categorize the new data that is
given as input. In dimensional view a hyperplane is a line that divide the data into different
categories. A line divide the data by which new data is classified.
Now as per the figure we can see a line divided the colors.Now let us consider a situation in
which a single line cannot separate the data forms. In this case we have to increase the
dimensions of the data. This can be done using many ways but radial basis functions is one of
best among them. After increasing the dimensions you can separate the data using a plane.
This plane acts as hyperplane and classify the new data accordingly. There is a possibility to
draw multiple hyperplanes in the machine. We have to select a hyperplane which is far from the
first point on both the sides of the classification. Thus maximizing the distance between the data
classes from the plane to find the optimal hyperplane.
There are some tuning factors in svm. These are
● Kernel
● Regulations
● Gamma
● Margin

You might also like