A class-based evaluation approach to assess multidimensional projections
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , , , |
Tipo de documento: | Artigo de conferência |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1109/IV51561.2020.00037 http://hdl.handle.net/11449/206083 |
Resumo: | Multidimensional projection techniques have been widely used to visually explore datasets due to their ability to generate representations that preserve similarity relations of data points into lower dimensional spaces. To evaluate if the embedded space reflects high-dimensional structures, measures are usually employed to return a quality score of the whole projection. In contrast to this idea, we evaluate the embedded layouts by assessing each class of the datasets at a time by using well-known quality measures. In addition, we propose assessing multidimensional projection techniques using ROC curves. Experimental results on two datasets show that our approach can be useful to discover how classes interact each other by using different visualization techniques and how close-related they are without thoroughly exploring the layouts. ROC curves proved to be a good measure for analyzing projection techniques and can give highly valuable feedback to users when exploring multidimensional data. |
id |
UNSP_c03b119f8d2ab008d61b241425de0d7b |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/206083 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
A class-based evaluation approach to assess multidimensional projectionsEvaluationMultidimensional projectionsVisualizationMultidimensional projection techniques have been widely used to visually explore datasets due to their ability to generate representations that preserve similarity relations of data points into lower dimensional spaces. To evaluate if the embedded space reflects high-dimensional structures, measures are usually employed to return a quality score of the whole projection. In contrast to this idea, we evaluate the embedded layouts by assessing each class of the datasets at a time by using well-known quality measures. In addition, we propose assessing multidimensional projection techniques using ROC curves. Experimental results on two datasets show that our approach can be useful to discover how classes interact each other by using different visualization techniques and how close-related they are without thoroughly exploring the layouts. ROC curves proved to be a good measure for analyzing projection techniques and can give highly valuable feedback to users when exploring multidimensional data.São Paulo State University (UNESP) Department of Mathematics and Computer ScienceDalhousie University Institute for Big Data AnalyticsSão Paulo State University (UNESP) Department of Mathematics and Computer ScienceUniversidade Estadual Paulista (Unesp)Institute for Big Data AnalyticsTeixeira, Jaqueline [UNESP]Marcilio, Wilson E. [UNESP]Eler, Danilo M. [UNESP]Artero, Almir [UNESP]Brandoli, Bruno2021-06-25T10:26:15Z2021-06-25T10:26:15Z2020-09-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject174-181http://dx.doi.org/10.1109/IV51561.2020.00037Proceedings of the International Conference on Information Visualisation, v. 2020-September, p. 174-181.1093-9547http://hdl.handle.net/11449/20608310.1109/IV51561.2020.000372-s2.0-85102927432Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengProceedings of the International Conference on Information Visualisationinfo:eu-repo/semantics/openAccess2021-10-22T20:56:10Zoai:repositorio.unesp.br:11449/206083Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T15:49:22.167201Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
A class-based evaluation approach to assess multidimensional projections |
title |
A class-based evaluation approach to assess multidimensional projections |
spellingShingle |
A class-based evaluation approach to assess multidimensional projections Teixeira, Jaqueline [UNESP] Evaluation Multidimensional projections Visualization |
title_short |
A class-based evaluation approach to assess multidimensional projections |
title_full |
A class-based evaluation approach to assess multidimensional projections |
title_fullStr |
A class-based evaluation approach to assess multidimensional projections |
title_full_unstemmed |
A class-based evaluation approach to assess multidimensional projections |
title_sort |
A class-based evaluation approach to assess multidimensional projections |
author |
Teixeira, Jaqueline [UNESP] |
author_facet |
Teixeira, Jaqueline [UNESP] Marcilio, Wilson E. [UNESP] Eler, Danilo M. [UNESP] Artero, Almir [UNESP] Brandoli, Bruno |
author_role |
author |
author2 |
Marcilio, Wilson E. [UNESP] Eler, Danilo M. [UNESP] Artero, Almir [UNESP] Brandoli, Bruno |
author2_role |
author author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) Institute for Big Data Analytics |
dc.contributor.author.fl_str_mv |
Teixeira, Jaqueline [UNESP] Marcilio, Wilson E. [UNESP] Eler, Danilo M. [UNESP] Artero, Almir [UNESP] Brandoli, Bruno |
dc.subject.por.fl_str_mv |
Evaluation Multidimensional projections Visualization |
topic |
Evaluation Multidimensional projections Visualization |
description |
Multidimensional projection techniques have been widely used to visually explore datasets due to their ability to generate representations that preserve similarity relations of data points into lower dimensional spaces. To evaluate if the embedded space reflects high-dimensional structures, measures are usually employed to return a quality score of the whole projection. In contrast to this idea, we evaluate the embedded layouts by assessing each class of the datasets at a time by using well-known quality measures. In addition, we propose assessing multidimensional projection techniques using ROC curves. Experimental results on two datasets show that our approach can be useful to discover how classes interact each other by using different visualization techniques and how close-related they are without thoroughly exploring the layouts. ROC curves proved to be a good measure for analyzing projection techniques and can give highly valuable feedback to users when exploring multidimensional data. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020-09-01 2021-06-25T10:26:15Z 2021-06-25T10:26:15Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/conferenceObject |
format |
conferenceObject |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1109/IV51561.2020.00037 Proceedings of the International Conference on Information Visualisation, v. 2020-September, p. 174-181. 1093-9547 http://hdl.handle.net/11449/206083 10.1109/IV51561.2020.00037 2-s2.0-85102927432 |
url |
http://dx.doi.org/10.1109/IV51561.2020.00037 http://hdl.handle.net/11449/206083 |
identifier_str_mv |
Proceedings of the International Conference on Information Visualisation, v. 2020-September, p. 174-181. 1093-9547 10.1109/IV51561.2020.00037 2-s2.0-85102927432 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Proceedings of the International Conference on Information Visualisation |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
174-181 |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808128567468359680 |