scholarly journals Rademacher Chaos Complexities for Learning the Kernel Problem

2010 ◽  
Vol 22 (11) ◽  
pp. 2858-2886 ◽  
Author(s):  
Yiming Ying ◽  
Colin Campbell
Keyword(s):  
Error Rates ◽  
Kernel Learning ◽  
Misclassification Error ◽  
Metric Entropy ◽  
Learning Problem ◽  
Generalization Bounds ◽  
Gaussian Kernels ◽  
Radial Basis ◽  
Generalization Bound ◽  
Chaos Process

We develop a novel generalization bound for learning the kernel problem. First, we show that the generalization analysis of the kernel learning problem reduces to investigation of the suprema of the Rademacher chaos process of order 2 over candidate kernels, which we refer to as Rademacher chaos complexity. Next, we show how to estimate the empirical Rademacher chaos complexity by well-established metric entropy integrals and pseudo-dimension of the set of candidate kernels. Our new methodology mainly depends on the principal theory of U-processes and entropy integrals. Finally, we establish satisfactory excess generalization bounds and misclassification error rates for learning gaussian kernels and general radial basis kernels.

Financial Fraud Detection using Radial Basis Network

2018 ◽  
Vol 3 (1) ◽  
pp. 10-21
Author(s):  
I. O. Alabi ◽  
R. G. Jimoh
Keyword(s):  
Radial Basis Function ◽  
Basis Function ◽  
Radial Basis Function Network ◽  
Fraud Detection ◽  
Misclassification Error ◽  
Computationally Efficient ◽  
Operating Characteristics ◽  
Data Mining Technique ◽  
Rbf Network ◽  
Radial Basis

The ubiquitous cases of abnormal transactions with intent to defraud is a global phenomenon. An architecture that enhances fraud detection using a radial basis function network was designed using a supervised data mining technique― radial basis function (RBF) network, a multivariate interpolation approximation method. Several base models were thus created, and in turn used in aggregation to select the optimum model using the misclassification error rate (MER), accuracy, sensitivity, specificity and receiver operating characteristics (ROC) metrics. The results shows that the model has a zero-tolerance for fraud with better prediction especially in cases where there were no fraud incidents doubtful cases were rather flagged than to allow a fraud incident to pass undetected. Expectedly, the model’s computations converge faster at 200 iterations. This study is generic with similar characteristics with other classification methods but distinct parameters thereby minimizing the time and cost of fraud detection by adopting computationally efficient algorithm.

The Performance of the Linear Discriminant Function in Nonoptimal Situations and the Estimation of Classification Error Rates: A Review of Recent Findings

1979 ◽  
Vol 16 (3) ◽  
pp. 370-381 ◽  
Author(s):  
William R. Dillon
Keyword(s):  
Discriminant Function ◽  
Error Rates ◽  
Linear Discriminant Function ◽  
Alternative Methods ◽  
Misclassification Error ◽  
Practical Significance ◽  
Classification Error ◽  
Continuous Variables ◽  
Discrete Variables ◽  
Linear Discriminant

This article is a review of the results, as are available, on the performance of the linear discriminant function in situations where the assumptions of multivariate normality and equal group dispersion structures are violated. Some new results are discussed for the case of classification using discrete variables, and in the case of both binary and continuous variables. In addition, alternative methods which have been proposed, and evaluated, for estimating misclassification error rates are thoroughly reviewed. In all cases, the material is reviewed in terms of practical significance, with particular emphasis on the conditions unfavorable to the performance of each procedure.

scholarly journals Peramalan Crude Palm Oil (CPO) Menggunakan Support Vector Regression Kernel Radial Basis

2017 ◽  
Vol 7 (1) ◽  
pp. 43 ◽  
Author(s):  
Rezzy Eko Caraka ◽  
Hasbi Yasin ◽  
Adi Waridi Basyiruddin
Keyword(s):  
Palm Oil ◽  
Oil Prices ◽  
Feature Space ◽  
Real Data ◽  
Kernel Learning ◽  
Support Vector ◽  
Data Sets ◽  
Crude Palm Oil ◽  
Machine Learning Methods ◽  
Radial Basis

Recently, instead of selecting a kernel has been proposed which uses SVR, where the weight of each kernel is optimized during training. Along this line of research, many pioneering kernel learning algorithms have been proposed. The use of kernels provides a powerful and principled approach to modeling nonlinear patterns through linear patterns in a feature space. Another bene?t is that the design of kernels and linear methods can be decoupled, which greatly facilitates the modularity of machine learning methods. We perform experiments on real data sets crude palm oil prices for application and better illustration using kernel radial basis. We see that evaluation gives a good to fit prediction and actual also good values showing the validity and accuracy of the realized model based on MAPE and R2. Keywords:  Crude Palm Oil; Forecasting; SVR; Radial Basis; Kernel

Error Estimations for Flow Characterization With Numerical and Analytical Solutions

2017 ◽  
Vol 6 (2) ◽  
pp. 16-28
Author(s):  
Irem Sanal
Keyword(s):  
Numerical Solution ◽  
3D Model ◽  
Error Rates ◽  
Basis Functions ◽  
Mean Square ◽  
Parallel Plates ◽  
Flow Characterization ◽  
Radial Basis ◽  
Potential Alternative ◽  
Error Estimations

Use of radial basis functions(RBFs) in the numerical solution of partial differential equations has gained popularity as it is meshless and can readily be extended to multi-dimensional problems. RBFs have been used in different context and emerged as a potential alternative for numerical solution of PDEs. In this article, a Flow Between Parallel Plates problem was solved using a Multiquadric Radial Basis Function Collocation Method (MQ-RBFCM), then, the results were compared with the analytic ones and the root mean square of the errors between the model and analytic results were calculated. Numerical results are presented for 5 different cases, where the number of inputs or definitions are increased to see whether changing the number of points makes the results better or not. Also, the absolute errors between the results were calculated to have a 3D model of the error rates and this has proven for which cases the MQ-RBFCM are better. As a result, RBF is shown to produce accurate results while requiring a much-reduced effort in problem preparation in comparison to traditional numerical methods.

scholarly journals Performance of Commercial Mycoplasma hyopneumoniae Serum Enzyme-Linked Immunosorbent Assays under Experimental and Field Conditions

2020 ◽  
Vol 58 (12) ◽  
Author(s):  
Ana Paula S. Poeta Silva ◽  
Ronaldo L. Magtoto ◽  
Henrique M. Souza Almeida ◽  
Aric McDaniel ◽  
Precy D. Magtoto ◽  
...  
Keyword(s):  
Serum Antibody ◽  
Mycoplasma Hyopneumoniae ◽  
Error Rates ◽  
False Positives ◽  
Field Conditions ◽  
Misclassification Error ◽  
Antibody Detection ◽  
Serum Samples ◽  
Content Type ◽  
Immunosorbent Assays

ABSTRACT Mycoplasma hyopneumoniae is an economically significant pathogen of swine. M. hyopneumoniae serum antibody detection via commercial enzyme-linked immunosorbent assays (ELISAs) is widely used for routine surveillance in commercial swine production systems. Samples from two studies were used to evaluate assay performance. In study 1, 6 commercial M. hyopneumoniae ELISAs were compared using serum samples from 8-week-old cesarean-derived, colostrum-deprived (CDCD) pigs allocated to the following 5 inoculation groups of 10 pigs each: (i) negative control, (ii) Mycoplasma flocculare (strain 27399), (iii) Mycoplasma hyorhinis (strain 38983), (iv) Mycoplasma hyosynoviae (strain 34428), and (v) M. hyopneumoniae (strain 232). Weekly serum and daily oral fluid samples were collected through 56 days postinoculation (dpi). The true status of pigs was established by PCR testing on oral fluids samples over the course of the observation period. Analysis of ELISA performance at various cutoffs found that the manufacturers’ recommended cutoffs were diagnostically specific, i.e., produced no false positives, with the exceptions of 2 ELISAs. An analysis based on overall misclassification error rates found that 4 ELISAs performed similarly, although one assay produced more false positives. In study 2, the 3 best-performing ELISAs from study 1 were compared using serum samples generated under field conditions. Ten 8-week-old pigs were intratracheally inoculated with M. hyopneumoniae. Matched serum and tracheal samples (to establish the true pig M. hyopneumoniae status) were collected at 7- to 14-day intervals through 98 dpi. Analyses of sensitivity and specificity showed similar performance among these 3 ELISAs. Overall, this study provides an assessment of the performance of current M. hyopneumoniae ELISAs and an understanding of their use in surveillance.

scholarly journals Comparing support vector machines with Gaussian kernels to radial basis function classifiers

1997 ◽  
Vol 45 (11) ◽  
pp. 2758-2765 ◽  
Author(s):  
B. Scholkopf ◽  
Kah-Kay Sung ◽  
C.J.C. Burges ◽  
F. Girosi ◽  
P. Niyogi ◽  
...  
Keyword(s):  
Support Vector Machines ◽  
Radial Basis Function ◽  
Basis Function ◽  
Support Vector ◽  
Gaussian Kernels ◽  
Vector Machines ◽  
Radial Basis ◽  
Radial Basis Function Classifiers

scholarly journals On the Error Rate Comparison of the Quadratic Discriminant Function, Euclidean Distance Classifier, Fisher’s Linear Discriminant Function and the Vine Copulas

2018 ◽  
Vol 8 (1) ◽  
pp. 113
Author(s):  
A. Nanthakumar
Keyword(s):  
Discriminant Function ◽  
Error Rate ◽  
Euclidean Distance ◽  
Error Rates ◽  
Linear Discriminant Function ◽  
Misclassification Error ◽  
Classification Problems ◽  
Linear Discriminant ◽  
Quadratic Discriminant Function ◽  
Vine Copulas

The estimation of the error rates is of vital importance in classification problems as this is used as a basis to choose the best discriminant function; that is, the one with a minimum misclassification error. The quadratic discriminant function (QDF), Euclidean Distance Classifier (EDC), and Fisher’s Linear Discriminant Function (FLDC) have been in use for a long time for the purpose of classification. In this paper, we compare the misclassification error rate of the QDF, EDC, and FLDC with the Vine Copulas based on Gaussian and Clayton models. The results were obtained for the general case where the means are unequal and the covariance matrices are unequal.

Quality Assessments of HMO Diagnosis Databases Used to Monitor Childhood Vaccine Safety

2004 ◽  
Vol 43 (02) ◽  
pp. 163-170 ◽  
Author(s):  
L. Drew ◽  
F. DeStefano ◽  
J. Maher ◽  
K. Bohlke ◽  
V. Immanuel ◽  
...  
Keyword(s):  
Medical Record ◽  
Vaccine Safety ◽  
Error Rates ◽  
Misclassification Error ◽  
Predictive Values ◽  
Clinical Criteria ◽  
Random Samples ◽  
Automated Screening ◽  
Assess Quality

Summary Objective: To assess the quality of automated diagnoses extracted from medical care databases by the Vaccine Safety Datalink (VSD) study. Methods: Two methods are used to assess quality of VSD diagnosis data. The first method compares common automated and abstracted diagnostic categories (“outcomes”) in 1-2% simple random samples of study populations. The second method estimates positive predictive values of automated diagnosis codes used to identify potential cases of rare conditions (e.g., acute ataxia) for inclusion in nested case-control medical record abstraction studies. Results: There was good agreement (64-68%) between automated and abstracted outcomes in the 1-2% simple random samples at 3 of the 4 VSD sites and poor agreement (44%) at 1 site. Overall at 3 sites, 56% of children with automated cerebella ataxia codes (ICD-9 = 334) and 22% with “lack of coordination” codes (ICD-9 = 781.3) met objective clinical criteria for acute ataxia. Conclusions: The misclassification error rates for automated screening outcomes substantially reduce the power of screening analyses and limit usefulness of screening analyses to moderate to strong vaccine-outcome associations. Medical record verification of outcomes is needed for definitive assessments.

Generalization Bounds of Regularization Algorithm with Gaussian Kernels

2013 ◽  
Vol 39 (2) ◽  
pp. 179-194 ◽  
Author(s):  
Feilong Cao ◽  
Yufang Liu ◽  
Weiguo Zhang
Keyword(s):  
Regularization Algorithm ◽  
Generalization Bounds ◽  
Gaussian Kernels
Sign in / Sign up

Export Citation Format

Share Document