Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
20 views9 pages

Advance Computer Architechture

Support Vector Machine (SVM) is a supervised learning algorithm used for classification and regression tasks, focusing on generating an optimal hyperplane to minimize errors. It utilizes support vectors to define the separating line and employs a kernel trick to handle non-linear separations by transforming data into higher dimensions. SVM is advantageous for its accuracy and efficiency with smaller datasets but struggles with large datasets and overlapping classes.

Uploaded by

alirehman123001
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
20 views9 pages

Advance Computer Architechture

Support Vector Machine (SVM) is a supervised learning algorithm used for classification and regression tasks, focusing on generating an optimal hyperplane to minimize errors. It utilizes support vectors to define the separating line and employs a kernel trick to handle non-linear separations by transforming data into higher dimensions. SVM is advantageous for its accuracy and efficiency with smaller datasets but struggles with large datasets and overlapping classes.

Uploaded by

alirehman123001
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Support vector

machine(SVM)
BY AMIR HATAM
Definition

 Kind of supervised learning.


 Used for classification as well as regression problems.
 Classification problem are those where the target contain
discrete no of possibilities.
 Regression problems, where target contain continuous values.
 SVM generates optimal hyper plane, used to minimize error.
 Core idea of SVM is to find a maximum marginal hyperplane that
best divides the data set into classes.
 Support vectors; data points which are closest to the hyper
plane. These points will define the separating line better by
calculating margins.
 Hyperplane; a decision plane which separates between a set
of objects having different class membership.
 Margin; a gap between the two lines on the closest class point.
 Its is perpendicular distance from line to support vectors.
How does SVM work
 Generates hyperplane which segregates the classes in the best
way.
 Select the right hyperplane with maximum segregation from the
either nearest data points.
 Its linear hyperplane.
Non linear hperplane

 Here SVM uses a kernel trick to transform the input space to a


higher dimensional space.
 A kernel transform an input data space into the required form
called kernel trick.
 Kernel trick convert non separable problems to separable by
adding more dimension to it
Tuning parameter

 1; Kernel , 2; regularization, 3; Gamma, 4; Margin


 Regularization; optimizes SVM classifier to avoid misclassifying
the data
 Gamma; how far influences the calculation of plausible line of
separation.
 Margin; perpendicular distance between the closest data points
and the hyperplane.
Application

 Face detection.
 Classification of images.
 Handwriting recognition etc.
Advantages
 Offer good occuracy and perform faster prediction.
 Used less memory because they used a subset of training points.
 Work with clear margin of separation and with dimensional
space.
Disadvantages
 Not suitable for large data set because of high training time.
 Works poorly with overlapping classes.
 Sensitive to the type of kernel used.

You might also like