SUPPORT

VECTOR
MACHINES
(SVMS)
Important
terms
What is the support vector ?

Support vectors
are the data
points that lie
closest to the
decision surface
(or hyperplane)
Why is this
called ?
It is called support because it supports me in
making the decision (to which class I belong) and
vector because it is treated like vectors in
mathematics.
What is the Hyperplane ?
hyperplane is a decision
boundary that differentiates
the two classes in SVM.
Is the Hyperplane
selection random,
and if not, how do I
determine?
It is possible to choose the
Hyperplane randomly, but this
will cause mix classification
 In general, lots of
possible solutions for
a ,b , c (an infinite
number!) , Support
Vector Machine (SVM)
finds an optimal
solution
 What are
support vector
machines
(SVMs)?
support vector machine is
a supervised learning
algorithm used to solve
classification and
regression problems.
How support vector
machines work ?
 A support vector machine (SVM) is a
supervised machine learning algorithm that
classifies data by finding an optimal line or
hyperplane that maximizes the distance
between each class in an N-dimensional
space.
(SVMs maximize the margin (Winston
terminology: the ‘street’) around the
separating hyperplane.)
After using Support Vector Machines
So how do we
calculate it
mathematically?
Form of equation defining the decision surface
separating the classes is a hyperplane of the form:

ax+b=o (linear)
wTx + b = 0
–w is a weight vector
–x is input vector
–b is bias
• Allows us to write
wTx + b ≥ 0 for di = +1
wTx + b < 0 for di = –1

H1 and H2 are the planes:

H1: w•xi+b = +1
H2: w•xi+b = –1
The points on the planes H1 and H2
are the tips of the Support Vectors
The plane H0 is the median in
between, where w•xi+b =0
 Define the hyperplanes H such that: w•xi+b ≥ +1 when yi =+1 & w•xi+b ≤ -1 when yi = –1

At H1
At H2
if w.x+b=1 then we get
if w.x+b=-1 then we get
(+)class hyperplane
(-)class hyperplane
for all positive(x) points
for all negative(x) points
At H0 satisfy this rule (w.x+b
satisfy this rule (w.x+b≤-
if w.x+b=0 then we get ≥1)
1)
the decision boundary
What if ?!
What if the
support
vector was
like this ?
In the previous example, the support vector cannot be separated linearly,
so we use kernal function, and this method is called non-linear
Thanks