Image analysis – Learning systems – Neural networks
Patent
1995-08-08
1998-08-18
Boudreau, Leo
Image analysis
Learning systems
Neural networks
382160, 382228, 395 23, G06T 140
Patent
active
057968632
ABSTRACT:
A statistical classifier is trained in a manner to remove biasing due to unequal frequencies of unigram priors. The relative frequencies of all classes in a training set of sample patterns is determined. Training patterns are then selected from the set and skipped or repeated in dependence upon the relative frequency of the class to which they belong. In this manner, the presentation of samples is balanced across the classes.
REFERENCES:
patent: 4479574 (1984-10-01), Glier et al.
patent: 4975975 (1990-12-01), Filipski
patent: 5325445 (1994-06-01), Herbert
patent: 5544257 (1996-08-01), Bellegarda et al.
patent: 5559929 (1996-09-01), Wasserman
Apple Computer Inc.
Boudreau Leo
Kelly Christoper S.
LandOfFree
Method for training an adaptive statistical classifier to balanc does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for training an adaptive statistical classifier to balanc, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for training an adaptive statistical classifier to balanc will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1123195