ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Outros Autores: | , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1016/j.bdr.2021.100239 http://hdl.handle.net/11449/221779 |
Resumo: | In exploratory tasks involving high-dimensional datasets, dimensionality reduction (DR) techniques help analysts to discover patterns and other useful information. Although scatter plot representations of DR results allow for cluster identification and similarity analysis, such a visual metaphor presents problems when the number of instances of the dataset increases, resulting in cluttered visualizations. In this work, we propose a scatter plot-based multilevel approach to display DR results and address clutter-related problems when visualizing large datasets, together with the definition of a methodology to use focus+context interaction on non-hierarchical embeddings. The proposed technique, called ExplorerTree, uses a sampling selection technique on scatter plots to reduce visual clutter and guide users through exploratory tasks. We demonstrate ExplorerTree's effectiveness through a use case, where we visually explore activation images of the convolutional layers of a neural network. Finally, we also conducted a user experiment to evaluate ExplorerTree's ability to convey embedding structures using different sampling strategies. |
id |
UNSP_d93d969255f39ab128bb40d79526270b |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/221779 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
ExplorerTree: A Focus+Context Exploration Approach for 2D EmbeddingsDimensionality reductionFocus+contextScatter-plotVisualizationIn exploratory tasks involving high-dimensional datasets, dimensionality reduction (DR) techniques help analysts to discover patterns and other useful information. Although scatter plot representations of DR results allow for cluster identification and similarity analysis, such a visual metaphor presents problems when the number of instances of the dataset increases, resulting in cluttered visualizations. In this work, we propose a scatter plot-based multilevel approach to display DR results and address clutter-related problems when visualizing large datasets, together with the definition of a methodology to use focus+context interaction on non-hierarchical embeddings. The proposed technique, called ExplorerTree, uses a sampling selection technique on scatter plots to reduce visual clutter and guide users through exploratory tasks. We demonstrate ExplorerTree's effectiveness through a use case, where we visually explore activation images of the convolutional layers of a neural network. Finally, we also conducted a user experiment to evaluate ExplorerTree's ability to convey embedding structures using different sampling strategies.Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Faculty of Sciences and Technology São Paulo State University (UNESP)Faculty of Computer Science Dalhousie UniversityInstitute of Mathematics and Computer Sciences University of São PauloFaculty of Sciences and Technology São Paulo State University (UNESP)FAPESP: 2016/11707-6FAPESP: 2017/17450-0FAPESP: 2018/17881-3FAPESP: 2018/25755-8Universidade Estadual Paulista (UNESP)Dalhousie UniversityUniversidade de São Paulo (USP)Marcílio-Jr, Wilson E. [UNESP]Eler, Danilo M. [UNESP]Paulovich, Fernando V.Rodrigues-Jr, José F.Artero, Almir O. [UNESP]2022-04-28T19:40:22Z2022-04-28T19:40:22Z2021-07-15info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1016/j.bdr.2021.100239Big Data Research, v. 25.2214-5796http://hdl.handle.net/11449/22177910.1016/j.bdr.2021.1002392-s2.0-85107938654Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengBig Data Researchinfo:eu-repo/semantics/openAccess2022-04-28T19:40:22Zoai:repositorio.unesp.br:11449/221779Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462022-04-28T19:40:22Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
title |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
spellingShingle |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings Marcílio-Jr, Wilson E. [UNESP] Dimensionality reduction Focus+context Scatter-plot Visualization |
title_short |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
title_full |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
title_fullStr |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
title_full_unstemmed |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
title_sort |
ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings |
author |
Marcílio-Jr, Wilson E. [UNESP] |
author_facet |
Marcílio-Jr, Wilson E. [UNESP] Eler, Danilo M. [UNESP] Paulovich, Fernando V. Rodrigues-Jr, José F. Artero, Almir O. [UNESP] |
author_role |
author |
author2 |
Eler, Danilo M. [UNESP] Paulovich, Fernando V. Rodrigues-Jr, José F. Artero, Almir O. [UNESP] |
author2_role |
author author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (UNESP) Dalhousie University Universidade de São Paulo (USP) |
dc.contributor.author.fl_str_mv |
Marcílio-Jr, Wilson E. [UNESP] Eler, Danilo M. [UNESP] Paulovich, Fernando V. Rodrigues-Jr, José F. Artero, Almir O. [UNESP] |
dc.subject.por.fl_str_mv |
Dimensionality reduction Focus+context Scatter-plot Visualization |
topic |
Dimensionality reduction Focus+context Scatter-plot Visualization |
description |
In exploratory tasks involving high-dimensional datasets, dimensionality reduction (DR) techniques help analysts to discover patterns and other useful information. Although scatter plot representations of DR results allow for cluster identification and similarity analysis, such a visual metaphor presents problems when the number of instances of the dataset increases, resulting in cluttered visualizations. In this work, we propose a scatter plot-based multilevel approach to display DR results and address clutter-related problems when visualizing large datasets, together with the definition of a methodology to use focus+context interaction on non-hierarchical embeddings. The proposed technique, called ExplorerTree, uses a sampling selection technique on scatter plots to reduce visual clutter and guide users through exploratory tasks. We demonstrate ExplorerTree's effectiveness through a use case, where we visually explore activation images of the convolutional layers of a neural network. Finally, we also conducted a user experiment to evaluate ExplorerTree's ability to convey embedding structures using different sampling strategies. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021-07-15 2022-04-28T19:40:22Z 2022-04-28T19:40:22Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1016/j.bdr.2021.100239 Big Data Research, v. 25. 2214-5796 http://hdl.handle.net/11449/221779 10.1016/j.bdr.2021.100239 2-s2.0-85107938654 |
url |
http://dx.doi.org/10.1016/j.bdr.2021.100239 http://hdl.handle.net/11449/221779 |
identifier_str_mv |
Big Data Research, v. 25. 2214-5796 10.1016/j.bdr.2021.100239 2-s2.0-85107938654 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Big Data Research |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
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_ |
1803649458841321472 |