Decision boundary feature extraction for nonparametric classification
IEEE Transactions on Systems Man and CyberneticsPublished 1 January 1993
C. Lee, D. A. Landgrebe
Citations52
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A new feature extraction algorithm based on decision boundaries for nonparametric classifiers is proposed, noting that feature extraction for pattern recognition is equivalent to retaining discriminantly informative features, and a discriminant informative feature is related to the decision boundary.
Abstract
A new feature extraction algorithm based on decision boundaries for nonparametric classifiers is proposed. It is noted that feature extraction for pattern recognition is equivalent to retaining discriminantly informative features, and a discriminantly informative feature is related to the decision boundary. Since nonparametric classifiers do not define decision boundaries in analytic form, the decision boundary and normal vectors must be estimated numerically. A procedure to extract discriminantly informative features based on a decision boundary for nonparametric classification is proposed. Experimental results show that the proposed algorithm finds effective features for the nonparametric classifier with Parzen density estimation.>
Keywords
Computer ScienceMathematics
Pattern classification and scene analysis
12,643 Citations1973Richard O. Duda, Peter E. Hart
Elsevier eBooksIntroduction to Statistical Pattern Recognition
11,121 Citations1990Keinosuke Fukunaga
A system to determine the sex of fish from their length, as accurately as possible, using a set of labelled data, together with the class label (male or female), as a training set.
IRE Transactions on Communications SystemsThe Divergence and Bhattacharyya Distance Measures in Signal Selection
1,844 Citations1967T. Kailath
This partly tutorial paper compares the properties of an often used measure, the divergence, with a new measure that is often easier to evaluate, called the Bhattacharyya distance, which gives results that are at least as good and often better than those given by the divergence.
IEEE Transactions on ComputersApplication of the Karhunen-Loève Expansion to Feature Selection and Ordering
537 Citations1970Keinosuke Fukunaga, Warren Koontz
A method is developed herein to use the Karhunen-Loeve expansion to extract features relevant to classification of a sample taken from one of two pattern classes.
IEEE Transactions on ComputersAn Optimal Set of Discriminant Vectors
404 Citations1975Donald H. Foley, John W. Sammon
A new method for the extraction of features in a two-class pattern recognition problem is derived that is based entirely upon discrimination or separability as opposed to the more common approach of fitting.
IEEE Transactions on Pattern Analysis and Machine IntelligenceFeature extraction based on decision boundaries
376 Citations1993C. Lee, D. A. Landgrebe
The proposed feature extraction algorithm has several desirable properties: it predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space for a given pattern recognition problem; and it finds the necessary feature vectors.
IEEE Transactions on Pattern Analysis and Machine IntelligenceNonparametric Discriminant Analysis
223 Citations1983Keinosuke Fukunaga, James M. Mantock
Using the same basic framework, a procedure is proposed to test the structural similarity of two distributions, and the form which results provides a unified view of the parametric nearest mean reclassification algorithm and the nonparametric valley seeking algorithm.
Introduction to Statistical Pattern Recognition
160 Citations2002Andrew R. Webb
Two approaches to dimensionality reduction, namely feature selection (FS) and feature extraction (FE) are specified, though FS is a special case of FE, they are very different from a practical viewpoint and thus must be considered separately.
IEEE Transactions on Pattern Analysis and Machine IntelligenceBayes Error Estimation Using Parzen and k-NN Procedures
129 Citations1987Keinosuke Fukunaga, D.M. Hummels
It is shown that, by varying the decision threshold, many of the biases associated with the k-NN or Parzen density estimates may be compensated, and successful error estimation may be performed in spite of these biases.
Purdue e-Pubs (Purdue University System)Two effective feature selection criteria for multispectral remote sensing
106 Citations1973P. H. Swain, Rachel King
The investigation provides an explanation for previous observations that the JM-distance and a saturating transform of divergence are highly useful for feature selection in the multiclass case.
IEEE Transactions on Information TheoryNonparametric feature selection
59 Citations1969E. Patrick, F.P. Fischer
A method is discussed for obtaining an l -dimensional linear subspace of the observation space in which the l -variate marginal distributions are most separated, based on a nonparametric estimate of probability density functions and a distance criterion.
Purdue e-Pubs (Purdue University)Feature extraction and classification algorithms for high-dimensional data
42 Citations1992Chulhee Lee, D. A. Landgrebe
A multistage classification scheme is proposed which reduces the processing time substantially by eliminating unlikely classes from further consideration at each stage by exploiting the increased importance of the second order statistics in analyzing high dimensional data.
Purdue e-Pubs (Purdue University)A crops and soils data base for scene radiation research
40 Citations1982L. L. Biehl, Marvin E. Bauer +4 more
IEEE Transactions on Geoscience and Remote SensingA spectral feature design system for the HIRIS/MODIS era
21 Citations1989C.-C.T. Chen, D. A. Landgrebe
A spectral feature design system for high-dimensional multispectral data that utilizes hard-limited or infinitely clipped optimal transforms and canonical analysis to extract the spectral features for data volume reduction and classification purposes is described.
IEEE Transactions on Geoscience and Remote SensingThe K-L Expansion as an Effective Feature Ordering Technique for Limited Training Sample Size
21 Citations1983M. J. Muasher, D. A. Landgrebe
Experimental results on the Hughes phenomenon using an effective feature ordering technique in situations where the number of training samples is limited in classifying multivariate two-class normal distributions are presented.
IEEE Transactions on ComputersGeneralized Clustering for Problem Localization
11 Citations1978Fukunaga, Kevin R. Short
The general procedure of conventional clustering is modified for various applications involving problem localization, and the concept of clustering criteria which are used for partitioning a training set, and depend upon certain a priori information with regards to the training set are introduced.
IEEE Transactions on Pattern Analysis and Machine IntelligenceFeature Extraction Using Problem Localization
7 Citations1982Robert D. Short, Keinosuke Fukunaga
Feature extraction is considered as a mean-quare estimation of the Bayes risk vector and partitioning the distribution space into local subregions and performing a linear estimation in each subregion minimizes the mean-square error.
Feature Selection For Neural Networks Using Parzen Density Estimator
5 Citations2005Chulhee Lee, Jón Atli Benediktsson +1 more
Using a reduced feature set, an attempt is made to reduce the training time of the neural network and obtain a simpler neural network, which further reduces the classification time for test data.