BMCBIOINFORMATICS . BIOMEDCENTRAL . COM {}

Analyzed Page
Matching Content Categories
CMS
Monthly Traffic Estimate
How Does Bmcbioinformatics.biomedcentral.com Make Money
How Much Does Bmcbioinformatics.biomedcentral.com Make
Keywords
Topics
Schema
Social Networks
External Links
Analytics And Tracking
Libraries
CDN Services

We are analyzing https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-12-138.

Title:
Elastic SCAD as a novel penalization method for SVM classification tasks in high-dimensional data | BMC Bioinformatics | Full Text
Description:
Background Classification and variable selection play an important role in knowledge discovery in high-dimensional data. Although Support Vector Machine (SVM) algorithms are among the most powerful classification and prediction methods with a wide range of scientific applications, the SVM does not include automatic feature selection and therefore a number of feature selection procedures have been developed. Regularisation approaches extend SVM to a feature selection method in a flexible way using penalty functions like LASSO, SCAD and Elastic Net. We propose a novel penalty function for SVM classification tasks, Elastic SCAD, a combination of SCAD and ridge penalties which overcomes the limitations of each penalty alone. Since SVM models are extremely sensitive to the choice of tuning parameters, we adopted an interval search algorithm, which in comparison to a fixed grid search finds rapidly and more precisely a global optimal solution. Results Feature selection methods with combined penalties (Elastic Net and Elastic SCAD SVMs) are more robust to a change of the model complexity than methods using single penalties. Our simulation study showed that Elastic SCAD SVM outperformed LASSO (L1) and SCAD SVMs. Moreover, Elastic SCAD SVM provided sparser classifiers in terms of median number of features selected than Elastic Net SVM and often better predicted than Elastic Net in terms of misclassification error. Finally, we applied the penalization methods described above on four publicly available breast cancer data sets. Elastic SCAD SVM was the only method providing robust classifiers in sparse and non-sparse situations. Conclusions The proposed Elastic SCAD SVM algorithm provides the advantages of the SCAD penalty and at the same time avoids sparsity limitations for non-sparse data. We were first to demonstrate that the integration of the interval search algorithm and penalized SVM classification techniques provides fast solutions on the optimization of tuning parameters. The penalized SVM classification algorithms as well as fixed grid and interval search for finding appropriate tuning parameters were implemented in our freely available R package
Website Age:
25 years and 10 months (reg. 1999-08-06).

Matching Content Categories {📚}

Science
Education
Telecommunications

Content Management System {📝}

What CMS is bmcbioinformatics.biomedcentral.com built with?

Custom-built

No common CMS systems were detected on Bmcbioinformatics.biomedcentral.com, and no known web development framework was identified.

Traffic Estimate {📈}

What is the average monthly size of bmcbioinformatics.biomedcentral.com audience?

🚀 Good Traffic: 50k - 100k visitors per month

Based on our best estimate, this website will receive around 53,076 visitors per month in the current month.

check SE Ranking
check Ahrefs
check Similarweb
check Ubersuggest
check Semrush

How Does Bmcbioinformatics.biomedcentral.com Make Money? {💸}

Display Ads {🎯}

The website utilizes display ads within its content to generate revenue. Check the next section for further revenue estimates.

Ads are managed by yourbow.com. Particular relationships are as follows:

Direct Advertisers (3)

yourbow.com, google.com, doceree.com

How Much Does Bmcbioinformatics.biomedcentral.com Make? {💰}

Display Ads {🎯}

$730 per month
Our estimates place Bmcbioinformatics.biomedcentral.com's monthly online earnings from display ads at $488 to $1,342.

Keywords {🔍}

svm, elastic, scad, data, features, selection, methods, feature, net, penalty, classification, set, google, scholar, tuning, sparse, classifier, classifiers, table, parameters, number, youden, index, prediction, article, showed, validation, vector, models, cancer, method, patients, search, error, cross, misclassification, performance, parameter, correlated, sensitivity, support, breast, rfe, positives, svms, specificity, algorithm, relevant, rate, simulation,

Topics {✒️}

/science/article/b6tbk-408bjcn-3/2/3a4753dc80ec448666ef990ee4c33078] 10 ten-fold stratified cross-validation support vector machine ten-fold cross validation author information authors maqc-ii breast cancer background classification k-fold cross validation references guyon support vector machines affymetrix hg-u133a platform sections receiver-operating characteristic analysis maqc-ii data set machine learning proceedings real high-dimensional data authors scientific editing stratified cross validation bmc bioinformatics 12 page gp expensive black-box functions fold cross validation authors’ original file full size image privacy choices/manage cookies lymph node negative 154 lymph node-negative springer nature high-dimensional prediction tasks machine learning 2002 feature selection svms called support vectors feature selection method automatic feature selection recursive feature elimination german federal ministry feature selection methods feature selection tasks input data vector small youden index feature selection procedures called leave elastic scad outperformed soft margin svm feature selection plays false positive results feature selection models scad svm provided youden index describes maximal youden index

Schema {🗺️}

WebPage:
      mainEntity:
         headline:Elastic SCAD as a novel penalization method for SVM classification tasks in high-dimensional data
         description:Classification and variable selection play an important role in knowledge discovery in high-dimensional data. Although Support Vector Machine (SVM) algorithms are among the most powerful classification and prediction methods with a wide range of scientific applications, the SVM does not include automatic feature selection and therefore a number of feature selection procedures have been developed. Regularisation approaches extend SVM to a feature selection method in a flexible way using penalty functions like LASSO, SCAD and Elastic Net. We propose a novel penalty function for SVM classification tasks, Elastic SCAD, a combination of SCAD and ridge penalties which overcomes the limitations of each penalty alone. Since SVM models are extremely sensitive to the choice of tuning parameters, we adopted an interval search algorithm, which in comparison to a fixed grid search finds rapidly and more precisely a global optimal solution. Feature selection methods with combined penalties (Elastic Net and Elastic SCAD SVMs) are more robust to a change of the model complexity than methods using single penalties. Our simulation study showed that Elastic SCAD SVM outperformed LASSO (L1) and SCAD SVMs. Moreover, Elastic SCAD SVM provided sparser classifiers in terms of median number of features selected than Elastic Net SVM and often better predicted than Elastic Net in terms of misclassification error. Finally, we applied the penalization methods described above on four publicly available breast cancer data sets. Elastic SCAD SVM was the only method providing robust classifiers in sparse and non-sparse situations. The proposed Elastic SCAD SVM algorithm provides the advantages of the SCAD penalty and at the same time avoids sparsity limitations for non-sparse data. We were first to demonstrate that the integration of the interval search algorithm and penalized SVM classification techniques provides fast solutions on the optimization of tuning parameters. The penalized SVM classification algorithms as well as fixed grid and interval search for finding appropriate tuning parameters were implemented in our freely available R package 'penalizedSVM'. We conclude that the Elastic SCAD SVM is a flexible and robust tool for classification and feature selection tasks for high-dimensional data such as microarray data sets.
         datePublished:2011-05-09T00:00:00Z
         dateModified:2011-05-09T00:00:00Z
         pageStart:1
         pageEnd:13
         license:http://creativecommons.org/licenses/by/2.0
         sameAs:https://doi.org/10.1186/1471-2105-12-138
         keywords:
            Support Vector Machine
            Feature Selection
            Feature Selection Method
            Youden Index
            Support Vector Machine Classification
            Bioinformatics
            Microarrays
            Computational Biology/Bioinformatics
            Computer Appl. in Life Sciences
            Algorithms
         image:
            https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig1_HTML.jpg
            https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig2_HTML.jpg
            https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig3_HTML.jpg
         isPartOf:
            name:BMC Bioinformatics
            issn:
               1471-2105
            volumeNumber:12
            type:
               Periodical
               PublicationVolume
         publisher:
            name:BioMed Central
            logo:
               url:https://www.springernature.com/app-sn/public/images/logo-springernature.png
               type:ImageObject
            type:Organization
         author:
               name:Natalia Becker
               affiliation:
                     name:INF 280
                     address:
                        name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                        type:PostalAddress
                     type:Organization
               email:[email protected]
               type:Person
               name:Grischa Toedt
               affiliation:
                     name:INF 280
                     address:
                        name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                        type:PostalAddress
                     type:Organization
               type:Person
               name:Peter Lichter
               affiliation:
                     name:INF 280
                     address:
                        name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                        type:PostalAddress
                     type:Organization
               type:Person
               name:Axel Benner
               affiliation:
                     name:INF 280
                     address:
                        name:German Cancer Research Center (DKFZ), Division Biostatistics, INF 280, Heidelberg, Germany
                        type:PostalAddress
                     type:Organization
               type:Person
         isAccessibleForFree:1
         type:ScholarlyArticle
      context:https://schema.org
ScholarlyArticle:
      headline:Elastic SCAD as a novel penalization method for SVM classification tasks in high-dimensional data
      description:Classification and variable selection play an important role in knowledge discovery in high-dimensional data. Although Support Vector Machine (SVM) algorithms are among the most powerful classification and prediction methods with a wide range of scientific applications, the SVM does not include automatic feature selection and therefore a number of feature selection procedures have been developed. Regularisation approaches extend SVM to a feature selection method in a flexible way using penalty functions like LASSO, SCAD and Elastic Net. We propose a novel penalty function for SVM classification tasks, Elastic SCAD, a combination of SCAD and ridge penalties which overcomes the limitations of each penalty alone. Since SVM models are extremely sensitive to the choice of tuning parameters, we adopted an interval search algorithm, which in comparison to a fixed grid search finds rapidly and more precisely a global optimal solution. Feature selection methods with combined penalties (Elastic Net and Elastic SCAD SVMs) are more robust to a change of the model complexity than methods using single penalties. Our simulation study showed that Elastic SCAD SVM outperformed LASSO (L1) and SCAD SVMs. Moreover, Elastic SCAD SVM provided sparser classifiers in terms of median number of features selected than Elastic Net SVM and often better predicted than Elastic Net in terms of misclassification error. Finally, we applied the penalization methods described above on four publicly available breast cancer data sets. Elastic SCAD SVM was the only method providing robust classifiers in sparse and non-sparse situations. The proposed Elastic SCAD SVM algorithm provides the advantages of the SCAD penalty and at the same time avoids sparsity limitations for non-sparse data. We were first to demonstrate that the integration of the interval search algorithm and penalized SVM classification techniques provides fast solutions on the optimization of tuning parameters. The penalized SVM classification algorithms as well as fixed grid and interval search for finding appropriate tuning parameters were implemented in our freely available R package 'penalizedSVM'. We conclude that the Elastic SCAD SVM is a flexible and robust tool for classification and feature selection tasks for high-dimensional data such as microarray data sets.
      datePublished:2011-05-09T00:00:00Z
      dateModified:2011-05-09T00:00:00Z
      pageStart:1
      pageEnd:13
      license:http://creativecommons.org/licenses/by/2.0
      sameAs:https://doi.org/10.1186/1471-2105-12-138
      keywords:
         Support Vector Machine
         Feature Selection
         Feature Selection Method
         Youden Index
         Support Vector Machine Classification
         Bioinformatics
         Microarrays
         Computational Biology/Bioinformatics
         Computer Appl. in Life Sciences
         Algorithms
      image:
         https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig1_HTML.jpg
         https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig2_HTML.jpg
         https://media.springernature.com/lw1200/springer-static/image/art%3A10.1186%2F1471-2105-12-138/MediaObjects/12859_2010_Article_4526_Fig3_HTML.jpg
      isPartOf:
         name:BMC Bioinformatics
         issn:
            1471-2105
         volumeNumber:12
         type:
            Periodical
            PublicationVolume
      publisher:
         name:BioMed Central
         logo:
            url:https://www.springernature.com/app-sn/public/images/logo-springernature.png
            type:ImageObject
         type:Organization
      author:
            name:Natalia Becker
            affiliation:
                  name:INF 280
                  address:
                     name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                     type:PostalAddress
                  type:Organization
            email:[email protected]
            type:Person
            name:Grischa Toedt
            affiliation:
                  name:INF 280
                  address:
                     name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                     type:PostalAddress
                  type:Organization
            type:Person
            name:Peter Lichter
            affiliation:
                  name:INF 280
                  address:
                     name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
                     type:PostalAddress
                  type:Organization
            type:Person
            name:Axel Benner
            affiliation:
                  name:INF 280
                  address:
                     name:German Cancer Research Center (DKFZ), Division Biostatistics, INF 280, Heidelberg, Germany
                     type:PostalAddress
                  type:Organization
            type:Person
      isAccessibleForFree:1
["Periodical","PublicationVolume"]:
      name:BMC Bioinformatics
      issn:
         1471-2105
      volumeNumber:12
Organization:
      name:BioMed Central
      logo:
         url:https://www.springernature.com/app-sn/public/images/logo-springernature.png
         type:ImageObject
      name:INF 280
      address:
         name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
         type:PostalAddress
      name:INF 280
      address:
         name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
         type:PostalAddress
      name:INF 280
      address:
         name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
         type:PostalAddress
      name:INF 280
      address:
         name:German Cancer Research Center (DKFZ), Division Biostatistics, INF 280, Heidelberg, Germany
         type:PostalAddress
ImageObject:
      url:https://www.springernature.com/app-sn/public/images/logo-springernature.png
Person:
      name:Natalia Becker
      affiliation:
            name:INF 280
            address:
               name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
               type:PostalAddress
            type:Organization
      email:[email protected]
      name:Grischa Toedt
      affiliation:
            name:INF 280
            address:
               name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
               type:PostalAddress
            type:Organization
      name:Peter Lichter
      affiliation:
            name:INF 280
            address:
               name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
               type:PostalAddress
            type:Organization
      name:Axel Benner
      affiliation:
            name:INF 280
            address:
               name:German Cancer Research Center (DKFZ), Division Biostatistics, INF 280, Heidelberg, Germany
               type:PostalAddress
            type:Organization
PostalAddress:
      name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
      name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
      name:German Cancer Research Center (DKFZ), Division Molecular Genetics, INF 280, Heidelberg, Germany
      name:German Cancer Research Center (DKFZ), Division Biostatistics, INF 280, Heidelberg, Germany

Social Networks {👍}(2)

External Links {🔗}(111)

Analytics and Tracking {📊}

Google Tag Manager

Libraries {📚}

Prism.js

CDN Services {📦}

Crossref

4.49s.