StAR: a simple tool for the statistical comparison of ROC curves
| dc.contributor.author | Vergara, Ismael A. | |
| dc.contributor.author | Norambuena Arenas, Tomás. | |
| dc.contributor.author | Ferrada, Evandro. | |
| dc.contributor.author | Slater Morales, Alex William. | |
| dc.contributor.author | Melo Ledermann, Francisco Javier | |
| dc.date.accessioned | 2019-10-17T18:19:08Z | |
| dc.date.available | 2019-10-17T18:19:08Z | |
| dc.date.issued | 2008 | |
| dc.date.updated | 2019-10-14T18:26:49Z | |
| dc.description.abstract | Abstract Background As in many different areas of science and technology, most important problems in bioinformatics rely on the proper development and assessment of binary classifiers. A generalized assessment of the performance of binary classifiers is typically carried out through the analysis of their receiver operating characteristic (ROC) curves. The area under the ROC curve (AUC) constitutes a popular indicator of the performance of a binary classifier. However, the assessment of the statistical significance of the difference between any two classifiers based on this measure is not a straightforward task, since not many freely available tools exist. Most existing software is either not free, difficult to use or not easy to automate when a comparative assessment of the performance of many binary classifiers is intended. This constitutes the typical scenario for the optimization of parameters when developing new classifiers and also for their performance validation through the comparison to previous art. Results In this work we describe and release new software to assess the statistical significance of the observed difference between the AUCs of any two classifiers for a common task estimated from paired data or unpaired balanced data. The software is able to perform a pairwise comparison of many classifiers in a single run, without requiring any expert or advanced knowledge to use it. The software relies on a non-parametric test for the difference of the AUCs that accounts for the correlation of the ROC curves. The results are displayed graphically and can be easily customized by the user. A human-readable report is generated and the complete data resulting from the analysis are also available for download, which can be used for further analysis with other software. The software is released as a web server that can be used in any client platform and also as a standalone application for the Linux operating system. Conclusion A new software for the statistical comparison of ROC curves is released here as a web server and also as standalone software for the LINUX operating system.Abstract Background As in many different areas of science and technology, most important problems in bioinformatics rely on the proper development and assessment of binary classifiers. A generalized assessment of the performance of binary classifiers is typically carried out through the analysis of their receiver operating characteristic (ROC) curves. The area under the ROC curve (AUC) constitutes a popular indicator of the performance of a binary classifier. However, the assessment of the statistical significance of the difference between any two classifiers based on this measure is not a straightforward task, since not many freely available tools exist. Most existing software is either not free, difficult to use or not easy to automate when a comparative assessment of the performance of many binary classifiers is intended. This constitutes the typical scenario for the optimization of parameters when developing new classifiers and also for their performance validation through the comparison to previous art. Results In this work we describe and release new software to assess the statistical significance of the observed difference between the AUCs of any two classifiers for a common task estimated from paired data or unpaired balanced data. The software is able to perform a pairwise comparison of many classifiers in a single run, without requiring any expert or advanced knowledge to use it. The software relies on a non-parametric test for the difference of the AUCs that accounts for the correlation of the ROC curves. The results are displayed graphically and can be easily customized by the user. A human-readable report is generated and the complete data resulting from the analysis are also available for download, which can be used for further analysis with other software. The software is released as a web server that can be used in any client platform and also as a standalone application for the Linux operating system. Conclusion A new software for the statistical comparison of ROC curves is released here as a web server and also as standalone software for the LINUX operating system.Abstract Background As in many different areas of science and technology, most important problems in bioinformatics rely on the proper development and assessment of binary classifiers. A generalized assessment of the performance of binary classifiers is typically carried out through the analysis of their receiver operating characteristic (ROC) curves. The area under the ROC curve (AUC) constitutes a popular indicator of the performance of a binary classifier. However, the assessment of the statistical significance of the difference between any two classifiers based on this measure is not a straightforward task, since not many freely available tools exist. Most existing software is either not free, difficult to use or not easy to automate when a comparative assessment of the performance of many binary classifiers is intended. This constitutes the typical scenario for the optimization of parameters when developing new classifiers and also for their performance validation through the comparison to previous art. Results In this work we describe and release new software to assess the statistical significance of the observed difference between the AUCs of any two classifiers for a common task estimated from paired data or unpaired balanced data. The software is able to perform a pairwise comparison of many classifiers in a single run, without requiring any expert or advanced knowledge to use it. The software relies on a non-parametric test for the difference of the AUCs that accounts for the correlation of the ROC curves. The results are displayed graphically and can be easily customized by the user. A human-readable report is generated and the complete data resulting from the analysis are also available for download, which can be used for further analysis with other software. The software is released as a web server that can be used in any client platform and also as a standalone application for the Linux operating system. Conclusion A new software for the statistical comparison of ROC curves is released here as a web server and also as standalone software for the LINUX operating system. | |
| dc.fuente.origen | Biomed Central | |
| dc.identifier.citation | BMC Bioinformatics. 2008 Jun 05;9(1):265 | |
| dc.identifier.doi | 10.1186/1471-2105-9-265 | |
| dc.identifier.uri | https://repositorio.uc.cl/handle/11534/26811 | |
| dc.issue.numero | No. 265 | |
| dc.language.iso | en | |
| dc.nota.acceso | Contenido completo | |
| dc.pagina.final | 5 | |
| dc.pagina.inicio | 1 | |
| dc.revista | BMC Bioinformatics | es_ES |
| dc.rights | acceso abierto | |
| dc.rights.holder | Vergara et al; licensee BioMed Central Ltd. | |
| dc.subject.ddc | 510 | |
| dc.subject.dewey | Matemática física y química | es_ES |
| dc.subject.other | Curvas algebraicas | es_ES |
| dc.subject.other | Matematicas | es_ES |
| dc.subject.other | Probabilidades | es_ES |
| dc.title | StAR: a simple tool for the statistical comparison of ROC curves | es_ES |
| dc.type | artículo | |
| dc.volumen | Vol.9 | |
| sipa.codpersvinculados | 73862 | |
| sipa.codpersvinculados | 121769 | |
| sipa.codpersvinculados | 82342 |