
	Where results make sense

Topic: Quadratic classifier

In the News (Mon 28 Dec 09)

Northern Trust

Marvin Harrison

Peter Chernin

Gary Locke

Match Play

Penelope Cruz

Mickey Rourke

AR Rahman

Danny Boyle

Hugh Jackman

	Quadratic classifier - Wikipedia, the free encyclopedia
		A quadratic classifier is used in machine learning to separate measurements of two or more classes of objects or events by a quadric surface.
		Finding a quadratic classifier for the original measurements would then become the same as finding a linear classifier based on the expanded measurement vector.
		For linear classifiers based only on dot products, these expanded measurements do not have to be actually computed, since the dot product in the higher dimensional space is simply related to that in the original space.
	en.wikipedia.org /wiki/Quadratic_classifier (411 words)

	A New Quadratic Classifier applied to Biometric Recognition (ResearchIndex)
		Abstract: In biometric recognition applications, the number of training examples per class is limited and consequently the conventional quadratic classifier either performs poorly or cannot be calculated.
		Other non-conventional quadratic classifiers have been used in limited sample and high dimensional classification problems.
		In this paper, a new quadratic classifier called Maximum Entropy Covariance Selection (MECS) is presented.
	citeseer.ist.psu.edu /691636.html (364 words)

	SAMPLEX 6 : Statistical pattern recognition classifier (Site not responding. Last check: 2007-10-11)
		The errors are the resubtitution (R error) and leave-one-out (L error) errors for the quadratic and linear Bayes classifiers.
		The resubstitution error is measured by classifying the database points (used to build the classifier) and counting the number of misclassification.
		The Rerror and Lerror are respectivelly the upper-bound and lower-bound error of the classifier.
	pasture.ecn.purdue.edu /~abe305/HTMLS/sample6.htm (201 words)

	Why active dendrites can remember more. (Site not responding. Last check: 2007-10-11)
		The memory capacity of an SQ classifier is given by the relation between the number of patterns to be learned and the error rate on an old-new discrimination task.
		We show that it is possible to characterize the performance of any SQ classifier in any dimension by scaling a reference curve in a reference dimension.
		In a neurobiological context, our results confirm that (1) a significant capacity boost is available to neurons whose dendrites provide even short-range multiplicative synaptic interactions, and (2) the spatial arrangement of synaptic contacts onto a dendritic tree could contain a large fraction of the neuron's memory-related state information.
	lnc.usc.edu /abstracts/98.sfn.poirazi_mel.html (305 words)

	Classifier Design
		The quadratic classifier will consequently be overparametrized, with the result that the training set will not be a good predictor for new cases.
		The resulting classifier is thus called a Gaussian linear classifier [81].
		In this procedure, the sample to be classified is withheld from the other samples, which are then used to design the classifier.
	splweb.bwh.harvard.edu:8000 /pages/papers/martin_thesis/node35.html (466 words)

	Examples: Statistical Pattern Recognition Toolbox for Matlab (Site not responding. Last check: 2007-10-11)
		The minimax approach is used to design a classifier prepared for the worst possible intervention.
		The figure shows quadratic classifier found by the Perceptron algorithm on the data mapped to the feature by the quadratic mapping.
		The class-conditional distributions are model by the Gaussian mixture models estimated by the EM algorithm.
	cmp.felk.cvut.cz /~xfrancv/stprtool/examples.html (506 words)

	Automatic discrimination of lung adenocarcinoma and pleura mesothelioma
		The quadratic classifier is very often successfully applied even if the patterns are not exact normally distributed.
		We applied a combination of multilayer perceptrons and quadratic classifiers to separate lung adenocarcinoma and plura mesothelioma tumour nuclei.
		These features and the neural network outputs were fed into the quadratic discriminant analysis and the class membership probabilities and the classification and generalization errors estimated.
	fuzzy.fzk.de /~rainer/new/work/node5.html (3745 words)

	311: Neural Networks (Site not responding. Last check: 2007-10-11)
		The quadratic Gaussian classifier also assumes multivariate normal distributions for each category, and that is why it is very similar to the linear classifier.
		The implication of having classes with different covariance matrices is that the decision boundary between the two classes is a quadratic hypersurface, that is a quadratic classifier:
		Training of a quadratic Gaussian classifier involves computing of the class mean vectors and covariance matrices.
	homepages.gold.ac.uk /nikolaev/311quadr.htm (220 words)

	COMPUTER ORGANIZATION (Site not responding. Last check: 2007-10-11)
		Draw a good configuration of the h points where the quadratic classifier is able to shatter perfectly and show the quadratic decision for all 2
		Also draw a poor configuration for h points that the quadratic classifier cannot shatter and show the particular labelings of the points that it cannot shatter.
		By ‘shatter’ we mean that the quadratic classifier will perfectly separate the data despite an arbitrary binary labeling of the points.
	www.cs.columbia.edu /~jebara/4771/HW2z.htm (984 words)

	Memory capacity model (Site not responding. Last check: 2007-10-11)
		However, the degree to which local nonlinear synaptic interactions augment the memory capacity of a neuron is not known in a quantitative sense.
		To approach this question, we have studied the family of subsampled quadratic classifiers, i.e.
		We developed an expression for the total parameter entropy of an SQ classifier, whose form shows that the capacity of an SQ classifier does not reside solely in its conventional weight values ---i.e.
	lnc.usc.edu /abstracts/98.poirazi_mel_memory.html (243 words)

	Applying artificial neural networks: Part II. Using near infrared data to classify tobacco types and identify native ...
		The models selected for this research were a quadratic classifier, a back-propagation network and a linear network.
		The results of the calibration model and the true performance for classifying tobacco species were (100%, 100%), (99.38%, 99.39%) and (95.19%, 99.26%) for the quadratic classifier, back-propagation network and linear network, respectively.
		The identification of native tobacco and its true performance were (100%, 100%) using a quadratic classifier, (89.12%, 88.46%) using a back-propagation network and (80.68%, 79.62%) using a linear network.
	www.nirpublications.com /abs/J05_0019.html (263 words)

	Technical Report # 104 (Site not responding. Last check: 2007-10-11)
		In preliminary comparisons, it was shown that when the optimal classification boundary that separated the two experimental categories was quadratic, the models provided roughly equivalent accounts of the data.
		In a pair of experiments in which the optimal classification boundary was of a more complex form than quadratic, the deterministic GCM significantly outperformed the GQC.
		Several decision bound models that postulated more complex decision boundaries than the GQC were developed and tested, including a connectionist learning model that implements quartic polynomial decision boundaries.
	www.cogs.indiana.edu /Publications/techreps1993/104.html (147 words)

	S (Site not responding. Last check: 2007-10-11)
		Unlike traditional source-filter theory, the CTF does not explicitly separate the spectral characteristics of the vocal source and the vocal tract filter.
		The principal components of the CTFs are used as features for a quadratic classifier to identify singers.
		The classifier's performance is not degraded when different vowels are included in classifier training and evaluation.
	home.earthlink.net /~bartscma/tsap.html (200 words)

	Uros Batricevic (Publications) (Site not responding. Last check: 2007-10-11)
		Abstract: In this paper we discuss quadratic classifier design, using statistics-based parametric approach and neural approach.
		It is shown how we can make a choice between linear and quadratic classifier analysing Bhattacharyya distance.
		We obtain more complex linear classifier by reparameterization of quadratic classifier.
	galeb.etf.bg.ac.yu /~uki/ukipb.html (285 words)

	On data depth and distribution-free discriminant analysis using separating surfaces, Anil K. Ghosh, Probal Chaudhuri
		For instance, a linear (or a quadratic) classifier finds the linear projection (or the quadratic function) of the measurement variables that will maximize the separation between the classes.
		Fisher's discriminant analysis, which is primarily motivated by the multivariate normal distribution, uses the first- and second-order moments of the training sample to build such classifiers.
		One of these classifiers is closely related to Tukey's half-space depth, while the other is based on the concept of regression depth.
	projecteuclid.org /Dienst/UI/1.0/Summarize/euclid.bj/1110228239 (1051 words)

	Statistical Pattern Recognition Toolbox for Matlab, Snapshots (Site not responding. Last check: 2007-10-11)
		The minimax approach is used to design a classifier prepared for the worst possible intervetion.
		The demo allows to interactivelly define a toy training sets and to train the SVM classifier using different kernels and regularization constants.
		The figure shows the Principal Component Anlysis used to find the 1D representation of the input 2D data with minal reconstruction error.
	cmp.felk.cvut.cz /~xfrancv/stprtool/snapshots.html (448 words)

	Style-Conscious Quadratic Field Classifier - Veeramachaneni, Fujisawa, Liu, Nagy (ResearchIndex)
		We present a method for exploiting such `style' consistency using a quadratic discriminant.
		We show that under reasonable assumptions on the feature and class distributions, the estimation of style parameters is simple and accurate.
		0.4: Adaptive Classifiers for Multi-Source OCR - Veeramachaneni, Nagy (2003)
	citeseer.ist.psu.edu /693940.html (443 words)

	[No title]
		In view of the fact that the variances are different, we attempt a quadratic classifier.
		Functions resub_crossval computes both the resubstitution and crossvalidation probabilities for the Fisher linear classifier, and requb_crossvalq does this for the quadratic classifier.
		The output values of the function are mat, a 3 index array that contains the three inverse covariance matrices, vecs which contains three vectors that are used for the linear portion of the quadratic classifier, and cons which has the three constants for the three discriminant function.
	www.ece.drexel.edu /courses/ECE-S680/Exams/estpar.doc (2302 words)

	List of Dr. C.J.Précetti Software (Site not responding. Last check: 2007-10-11)
		It was found more reliable and accurate than the quadratic Bayes classifier in the case of biomaterial color classification.
		Purclas (Purdue color classifier...) is a neural network based color classifier.
		It allows the user to sample different color classes from a 3 files RGB image, build a color classifier with either Bayes classifier or neural networks, classify the image, and display the result.
	pasture.ecn.purdue.edu /~abe305/HTMLS/mysoft2.htm (513 words)

	[No title] (Site not responding. Last check: 2007-10-11)
		The task is to generate some artificial data, train a few classifiers on it, estimate the error of the classifiers and make a plot of their decision boundaries.
		Next, we train two classifiers on the training set, a quadratic classifier and a k-nearest neighbor classifier (with k=3).
		There are actually two versions of the quadratic optimizer compiled for Windows.
	130.161.42.18 /prtools/documentation.html (341 words)

	Scalable Mining For Classification Rules in Relational Databases
		The objective of the classification is to process the DETAIL table and produce a classifier, which contains a description(model) for each class.
		The models will be used to classify future data for which the class labels are unknown.
		Experimental Results show very clearly MIND superiority over SPRINT (which is the state of the art classification algorithm known until know) when DETAIL table size is larger then one million records (over the authors Database).
	www.math.tau.ac.il /~matias/courses/sem_fall99/grossaug-summary.html (796 words)

	Index for Directory export/lyngby
		of regularization for weights lyngby_nn_cdev lyngby_nn_cdev - Classifier neural network, input 1st der.
		lyngby_nn_qddevds lyngby_nn_qddevds - 2nd der., quadratic, input, diag gauss.
		lyngby_nn_qddevs lyngby_nn_qddevs - 2nd der., quadratic, input, gauss approx.
	hendrix.imm.dtu.dk /software/lyngby/doc/lyngby.mat2html/1.lyngby (298 words)

	Liu
		Study of recent advances in development of statistical pattern recognition algorithms, approximation, and estimation techniques.
		), classifier design, parameter estimation, feature extraction (for representation and classification), clustering
		Quadratic Discriminant Analysis (QDA) and Linear Discriminant Analysis (LDA)
	www.cs.njit.edu /liu/Courses/2006Spring/CS782.html (203 words)

	[No title]
		Study of recent advances in development of (statistical and syntactic) pattern recognition algorithms, approximation, and estimation techniques.
		Topics include statistical estimation theory, classifier design, parameter estimation and unsupervised learning, bias vs.
		variance, nonparametric techniques, linear discriminant functions, tree classifiers, feature extraction, and clustering.
	www.cs.njit.edu /~liu/Courses/2003Spring/cs782.html (144 words)

	Design and Application of Quadratic Correlation Filters for Target Detection (Site not responding. Last check: 2007-10-11)
		Design and Application of Quadratic Correlation Filters for Target Detection
		We introduce a method for designing and implementing quadratic correlation filters (QCFs) for shift-invariant target detection in imagery.
		The QCFs are a quadratic classifier that operates directly on the image data without feature extraction or segmentation.
	www.ewh.ieee.org /soc/aes/taes/aes403/4030837.htm (238 words)

	Contents: Decision, Estimation & Classification (Site not responding. Last check: 2007-10-11)
		6.1 Quadratic and Linear Classifiers for Two-Class Problems
		10.2 A Recursive Form for the Quadratic Classifier
		10.4 Viewing the Recursive Form of the Classifier as a Causal Whitening Process
	web.nps.navy.mil /~therrien/dec.html (159 words)

Try your search on: Qwika (all wikis)