Results 1  10
of
28
Statistical pattern recognition: A review
 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
, 2000
"... The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques ..."
Abstract

Cited by 1035 (30 self)
 Add to MetaCart
The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques and methods imported from statistical learning theory have bean receiving increasing attention. The design of a recognition system requires careful attention to the following issues: definition of pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design and learning, selection of training and test samples, and performance evaluation. In spite of almost 50 years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition, and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this review paper is to summarize and compare some of the wellknown methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.
LargeScale Simulation Studies in Image Pattern Recognition
 IEEE Trans. on PAMI
, 1997
"... Many obstacles to progress in image pattern recognition result from the fact that perclass distributions are often too irregular to be wellapproximated by simple analytical functions. Simulation studies offer one way to circumvent these obstacles. We present three closely related studies of machin ..."
Abstract

Cited by 29 (10 self)
 Add to MetaCart
(Show Context)
Many obstacles to progress in image pattern recognition result from the fact that perclass distributions are often too irregular to be wellapproximated by simple analytical functions. Simulation studies offer one way to circumvent these obstacles. We present three closely related studies of machineprinted character recognition that rely on synthetic data generated pseudorandomly in accordance with an explicit stochastic model of document image degradations. The unusually large scale of experiments  involving several millions of samples  that this methodology makes possible has allowed us to compute sharp estimates of the intrinsic difficulty (Bayes risk) of concrete image recognition problems, as well as the asymptotic accuracy and domain of competency of classifiers. 1 Introduction In most of the literature on pattern recognition, image data sets, used to train and test classifiers, constitute the only specification of the problem offered. These image data are usually unava...
Learning pattern classification  A survey
 IEEE TRANS. INFORM. THEORY
, 1998
"... Classical and recent results in statistical pattern recognition and learning theory are reviewed in a twoclass pattern classification setting. This basic model best illustrates intuition and analysis techniques while still containing the essential features and serving as a prototype for many applic ..."
Abstract

Cited by 20 (4 self)
 Add to MetaCart
Classical and recent results in statistical pattern recognition and learning theory are reviewed in a twoclass pattern classification setting. This basic model best illustrates intuition and analysis techniques while still containing the essential features and serving as a prototype for many applications. Topics discussed include nearest neighbor, kernel, and histogram methods, Vapnik–Chervonenkis theory, and neural networks. The presentation and the large (thogh nonexhaustive) list of references is geared to provide a useful overview of this field for both specialists and nonspecialists.
Joint sampling distribution between actual and estimated classification errors for linear discriminant analysis
 IEEE Trans. Inf. Theory
, 2010
"... Abstract—Error estimation must be used to find the accuracy of a designed classifier, an issue that is critical in biomarker discovery for disease diagnosis and prognosis in genomics and proteomics. This paper presents, for what is believed to be the first time, the analytical formulation for the jo ..."
Abstract

Cited by 18 (9 self)
 Add to MetaCart
(Show Context)
Abstract—Error estimation must be used to find the accuracy of a designed classifier, an issue that is critical in biomarker discovery for disease diagnosis and prognosis in genomics and proteomics. This paper presents, for what is believed to be the first time, the analytical formulation for the joint sampling distribution of the actual and estimated errors of a classification rule. The analysis presented here concerns the linear discriminant analysis (LDA) classification rule and the resubstitution and leaveoneout error estimators, under a general parametric Gaussian assumption. Exact results are provided in the univariate case, and a simple method is suggested to obtain an accurate approximation in the multivariate case. It is also shown how these results can be applied in the computation of condition bounds and the regression of the actual error, given the observed error estimate. In contrast to asymptotic results, the analysis presented here is applicable to finite training data. In particular, it applies in the smallsample settings commonly found in genomics and proteomics applications. Numerical examples, which include parameters estimated from actual microarray data, illustrate the analysis throughout. Index Terms—Classification, crossvalidation, error estimation, leaveoneout, linear discriminant analysis, resubstitution, sampling distribution. I.
Fads and Fallacies in the Name of Smallsample Microarray Classification
 IEEE Signal Processing Magazine
, 2007
"... In his landmark book Fads and Fallacies In The Name of Science, Martin Gardner wrote the following words on pseudoscience [10, p. 15]: “The amount of lost energy that has been wasted on these lost causes is almost unbelievable. It will be amusing — at times frightening — to witness the grotesque ext ..."
Abstract

Cited by 16 (4 self)
 Add to MetaCart
(Show Context)
In his landmark book Fads and Fallacies In The Name of Science, Martin Gardner wrote the following words on pseudoscience [10, p. 15]: “The amount of lost energy that has been wasted on these lost causes is almost unbelievable. It will be amusing — at times frightening — to witness the grotesque extremes to which deluded scientists can be misled, and the extremes to
Data Complexity Analysis for Classifier Combination
 Proc. Int. Workshop on Multiple Classifier Systems (LNCS 2096
, 2001
"... Multiple classifier methods are effective solutions to difficult pattern recognition problems. However, empirical successes and failures have not been completely explained. Amid the excitement and confusion, uncertainty persists in the optimality of method choices for specific problems due to st ..."
Abstract

Cited by 13 (0 self)
 Add to MetaCart
(Show Context)
Multiple classifier methods are effective solutions to difficult pattern recognition problems. However, empirical successes and failures have not been completely explained. Amid the excitement and confusion, uncertainty persists in the optimality of method choices for specific problems due to strong data dependences of classifier performance. In response to this, I propose that further exploration of the methodology be guided by detailed descriptions of geometrical characteristics of data and classifier models.
Estimating the Posterior Probabilities Using the KNearest Neighbor Rule
, 2004
"... In many pattern classification problems an estimate of the posterior probabilities (rather than only a classification) is required. This is usually the case when some confidence measure in the classification is needed. In this paper we propose a new posterior probability estimator. The proposed est ..."
Abstract

Cited by 11 (0 self)
 Add to MetaCart
(Show Context)
In many pattern classification problems an estimate of the posterior probabilities (rather than only a classification) is required. This is usually the case when some confidence measure in the classification is needed. In this paper we propose a new posterior probability estimator. The proposed estimator considers the Knearest neighbors. It attaches a weight to each neighbor that contributes in an additive fashion to the posterior probability estimate. The weights corresponding to the Knearestneighbors (which add to 1) are estimated from the data using a maximum likelihood approach. Simulation studies confirm the effectiveness of the proposed estimator. 1
A Simple Method to Predict Protein Binding From Aligned Sequences
, 2005
"... Motivation: The MHC superfamily (MhcSF) consists of immune system MHC class I (MHCI) proteins, along with proteins with a MHCIlike structure that are involved in a large variety of biological processes. Beta2microglobulin (B2M) noncovalent binding to MHCI proteins is required for their surface e ..."
Abstract

Cited by 5 (5 self)
 Add to MetaCart
Motivation: The MHC superfamily (MhcSF) consists of immune system MHC class I (MHCI) proteins, along with proteins with a MHCIlike structure that are involved in a large variety of biological processes. Beta2microglobulin (B2M) noncovalent binding to MHCI proteins is required for their surface expression and function, while MHCIlike proteins interact, or not, with B2M. This study was designed to predict B2M binding (or nonbinding) of newly identified MhcSF proteins, in order to decipher their function, understand the molecular recognition mechanisms, and identify deleterious mutations. IMGT standardization of MhcSF protein domains provides a unique numbering of the multiple alignment positions, and conditions to develop such predictive tool.
Unsupervised supervised learning I: Estimating classification and regression error rates without labels
 Journal of Machine Learning Research
"... Estimating the error rates of classifiers or regression models is a fundamental task in machine learning which has thus far been studied exclusively using supervised learning techniques. We propose a novel unsupervised framework for estimating these error rates using only unlabeled data and mild ass ..."
Abstract

Cited by 5 (2 self)
 Add to MetaCart
(Show Context)
Estimating the error rates of classifiers or regression models is a fundamental task in machine learning which has thus far been studied exclusively using supervised learning techniques. We propose a novel unsupervised framework for estimating these error rates using only unlabeled data and mild assumptions. We prove consistency results for the framework and demonstrate its practical applicability on both synthetic and real world data.
Estimating the Intrinsic Difficulty of A Recognition Problem
 Proceedings of the 12th International Conference on Pattern Recognition
, 1994
"... We describe an experiment in estimating the Bayes error of a concrete image classification problem: a difficult, practically important, twoclass character recognition problem. The Bayes error gives the "intrinsic difficulty" of the problem since it is the minimum error achievable by any ..."
Abstract

Cited by 4 (1 self)
 Add to MetaCart
(Show Context)
We describe an experiment in estimating the Bayes error of a concrete image classification problem: a difficult, practically important, twoclass character recognition problem. The Bayes error gives the "intrinsic difficulty" of the problem since it is the minimum error achievable by any classification method. Since for many realistically complex problems, deriving this analytically appears to be hopeless, we approach the task empirically. We proceed first by expressing the problem precisely in terms of ideal prototype images and an image defect model, and then by carrying out the estimation on pseudorandomly simulated data. Arriving at sharp estimates seems inevitably to require both large sample sizes  in our trial, over a million images  and careful statistical extrapolation. The study of the data reveals many interesting statistics, which allow the prediction of the worstcase time/space requirements for any given classifier performance, expressed as a combination of error ...