Probability estimate for K-nearest neighbor

Details Probability estimate for K-nearest neighbor Probability estimate for K-nearest neighbor

2005-12-08

2008-11-11

Vincent, David (Department: 2129)

Data processing: artificial intelligence

Neural network

Learning task

C706S025000, C706S048000

Reexamination Certificate

active

07451123

ABSTRACT:
Systems and methods are disclosed that facilitate producing probabilistic outputs also referred to as posterior probabilities. The probabilistic outputs include an estimate of classification strength. The present invention intercepts non-probabilistic classifier output and applies a set of kernel models based on a softmax function to derive the desired probabilistic outputs. Such probabilistic outputs can be employed with handwriting recognition where the probability of a handwriting sample classification is combined with language models to make better classification decisions.

REFERENCES:
patent: 5594807 (1997-01-01), Liu
patent: 5880738 (1999-03-01), Donelly
patent: 5903884 (1999-05-01), Lyon et al.
patent: 5943661 (1999-08-01), Katz
patent: 6003027 (1999-12-01), Prager
patent: 6038337 (2000-03-01), Lawrence et al.
patent: 6061472 (2000-05-01), Hullender et al.
patent: 6151424 (2000-11-01), Hsu
patent: 6192360 (2001-02-01), Dumais et al.
patent: 6233575 (2001-05-01), Agrawal et al.
patent: 6324532 (2001-11-01), Spence et al.
patent: 6327581 (2001-12-01), Platt
patent: 6347309 (2002-02-01), De Tremiolles et al.
patent: 6519368 (2003-02-01), Kondo et al.
patent: 6629937 (2003-10-01), Watrous
patent: 6728690 (2004-04-01), Meek et al.
patent: 6732083 (2004-05-01), Bax
patent: 6782373 (2004-08-01), de Tremiolles et al.
patent: 6895338 (2005-05-01), Hsiung et al.
patent: 6973215 (2005-12-01), Hullender et al.
patent: 7003158 (2006-02-01), Bennett et al.
patent: 7107254 (2006-09-01), Dumais et al.
patent: 2002/0150295 (2002-10-01), Kwok et al.
A. Tuerk and S.J. Young, “Polynomial Softmax Functions for Pattern Classification”, Cambridge University Engineering Department, Feb. 2001, 31 pages.
Eric B. Baum and Frank Wilczek, “Supervised Learning of Probability Distributions by Neural Networks”, Institute for Theoretical Physics, University of California, 1988, 10 pages.
Christopher C. Burges and David J. Crisp, “Uniqueness of the SVM Solution”, 2000, 7 pages.
Morris H. Degroot and Stephen E. Fienberg, “Assessing probability Assessors: Calibration and Refinement”, Department of Statistics Carnegie-Mellon University, 1982, 13 pages.
John H. Miller, Rod Goodman, Padhraic Smyth, “On Loss Functions Which Minimize to Conditional Expected Values and Posterior Probabilities”, IEEE Transactions on Information Theory, vol. 39, No. 4, Jul. 1993, 5 pages.
John S. Bridle, “Training Stochastic Model Recognition Algorithms as Networks can lead to Maximum Mutual Information Estimation of Parameters”, Royal Signals and Radar Establishment, 1990, 7 pages.
Allan H. Murray and Robert L. Winkler, “A General Framework for Forecast Verification”, Monthly Weather Review, vol. 15, Dec. 18, 1996, 9 pages.
John C. Platt, “Probabilities for SV Machines”, Microsoft Research, 2000, 13 pages.
T.J. Hastie and R.J. Tibshirani, Generalized Additive Models, Monographs on Statistics and Applied Probability, 1990, 335 pages.
R. Fletcher, “Practical Methods of Optimization”, Second Edition, 1987, 436 pages.
Robert P.W. Duin and David M.J. Tax, “Classifier Conditional Posterior Probabilities”, Aug. 1998, 9 pages.
David D. Lewis and Jason Catlett, “Heterogeneous uncertainty Sampling for Supervised Learning”, 9 pages.

Affiliated with

Also associated with

Rating

Probability estimate for K-nearest neighbor does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Probability estimate for K-nearest neighbor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Probability estimate for K-nearest neighbor will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4032898