Fast method for training and evaluating support vector...

Details Fast method for training and evaluating support vector... Fast method for training and evaluating support vector...

2007-02-06

2007-02-06

Chawan, Sheela (Department: 2624)

Image analysis

Learning systems

C382S224000, C382S253000, C382S276000, C706S020000

Reexamination Certificate

active

10199744

ABSTRACT:
A procedure for fast training and evaluation of support vector machines (SVMs) with linear input features of high dimensionality is presented. The linear input features are derived from raw input data by means of a set of m linear functions defined on the k-dimensional raw input data. Training uses a one-time precomputation on the linear transform matrix in order to allow training on an equivalent training set with vector size k instead of m, given a great computational benefit in case of m>>k. A similar precomputation is used during evaluation of SVMs, so that the raw input data vector can be used instead of the derived linear feature vector.

REFERENCES:
patent: 5950146 (1999-09-01), Vapnik
patent: 6112195 (2000-08-01), Burges
patent: 6134344 (2000-10-01), Burges
patent: 6327581 (2001-12-01), Platt
patent: 6658395 (2003-12-01), Barnhill
patent: 6671391 (2003-12-01), Zhang et al.
patent: 6803933 (2004-10-01), Staelin et al.
patent: 6882990 (2005-04-01), Barnhill et al.

Affiliated with

Also associated with

Rating

Fast method for training and evaluating support vector... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Fast method for training and evaluating support vector..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fast method for training and evaluating support vector... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3837606