submit urlsubmit rss feedadd directory

article

Support vector machines (SVMs) are a set of related supervised learning methods used for classification and regression. Their common factor is the use of a technique known as the "kernel trick" to apply linear classification techniques to non-linear classification problems.

Linear classification


Motivation

Suppose we want to classify some data points into two classes. Often we are interested in classifying data as part of a machine-learning process. These data points may not necessarily be points in R2 but may be multidimensional Rp (statistics notation) or Rn (computer science notation) points. We are interested in whether we can separate them by a hyperplane. As we examine a hyperplane, this form of classification is known as linear classification. We also want to choose a hyperplane that separates the data points "neatly", with maximum distance to the closest data point from both classes -- this distance is called the margin. We desire this property since if we add another data point to the points we already have, we can more accurately classify the new point since the separation between the two classes is greater. Now, if such a hyperplane exists, the hyperplane is clearly of interest and is known as the maximum-margin hyperplane or the optimal hyperplane, as are the vectors that are closest to this hyperplane, which are called the support vectors.

Formalization

We consider data points of the form: \{ (\mathbf{x}_1, c_1), (\mathbf{x}_2, c_2), \ldots, (\mathbf{x}_n, c_n)\} where the ci is either 1 or −1 -- this constant denotes the class to which the point \mathbf{x}_i belongs. Each \mathbf{x}_i is a p- (statistics notation), or n- (computer science notation) dimensional vector of scaled or [-1,1 values. The scaling is important to guard against variables (attributes) with larger variance that might otherwise dominate the classification. We can view this as training data, which denotes the correct classification which we would like the SVM to eventually distinguish, by means of the dividing hyperplane, which takes the form
\mathbf{w}\cdot\mathbf{x} - b=0.

More on [ Support vector machine ]


directory of related categories

 

 
Support_Vector_Machines RSS feed
MIT News: Robotics / artificial intelligence

MIT's new underwater robot can hover in place
Thu, 25 Sep 2008 00:00:00 -0500
MIT researchers have designed the Odyssey IV, a new robotic underwater vehicle that can hover in place like a helicopter and serve as an invaluable tool for deepwater oil explorers, marine archaeologists, oceanographers and others.
Robot wheelchair finds its own way
Fri, 19 Sep 2008 00:00:00 -0500
MIT researchers are developing a new kind of autonomous wheelchair that can learn all about the locations in a given building, and then take its occupant to a given place in response to a verbal command.
Memory capacity bigger than previously thought
Mon, 08 Sep 2008 00:00:00 -0500
A new study from MIT cognitive neuroscientists may overturn the widespread belief that human memory does not store the details of our experiences. They have shown, given the right setting, the human brain can record an amazing amount of information.
Model helps computers sort data more like humans
Mon, 25 Aug 2008 00:00:00 -0500
In an advance that may impact the field of artificial intelligence, a new model developed at MIT can help computers recognize patterns the same way that humans do. The model can analyze a set of data and figure out which type of organizational structure best fits it.
MIT class asks: Fly me to the moon?
Wed, 21 May 2008 00:00:00 -0500
An MIT graduate class, aimed at figuring out whether MIT could, or should, mount an entry into the $20-million Google Lunar X-Prize competition announced last fall, has arrived at the bottom line: Yes, we can (technically)!
Beaver-like robots face off in annual MIT contest
Fri, 09 May 2008 00:00:00 -0500
Robots designed to toss pool-noodle trees into a river of ping-pong balls ruled over competitors focused on rescuing fuzzy toy beavers in this year's 2.007 contest, "Da (yes) MIT, or Save the Baby Beavers," held on Thursday, May 8, at MIT.

Latest Issue of Journal of Experimental & Theoretical Artificial Intelligence

Introduction to the special issue on: pluralism and the future of cognitive science
Dale, Rick
The possibility of a pluralist cognitive science
Dale, Rick
On the nature of minds, or: truth and consequences
Edelman, Shimon

 
Subscribe to Artificial_Intelligence RSS feed

directory of related sites

artificial intelligence using SVM - Helpful to beginners trying to grasp the concepts of SVMs.

Image, Speech and Intelligent Systems Research Group - University of Southampton. Overview and links to resources.

Kernel Machines - A central source of information on kernel based methods, including support vector machines, Gaussian processes.

Lagrangian Support Vector Machine - University of Wisconsin at Madison. Software and technical report.
Meta Description: [ Active Support Vector Machine Home page ]

Learning to Classify Text using Support Vector Machines - By Thorsten Joachims - describes an SVM approach to text classification.

Support Vector Machine Mailing List - An unmoderated discussion list about Support Vector Machines methodology.

SVM Application List - Overview of domains in which SVMs have been applied.

Support_Vector_Machines related videos

Guys in the lab

This is the team of nerds of the Support Vector Machine project at CSUN. We're 4 grad students and an ugrad. It's good times

 

HOMEADVERTISINGABOUT US

articlesartsbusinesscomputersgameshealthhospitalshomekids & teensnewsmobilephysiciansrecreationreferenceregionalscienceshoppingsocietysportsworld


Submit a Site About Become an Editor