Mapping without visualizing local default is nonsense

Sylvain Lespinats, Michael Aupetit

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

High-dimensional data sets are often embedded in two-dimensional spaces so as to visualize neighborhood relationships. When the map is effective (i.e. when short distances are preserved) it is a powerful way to help an analyst to understand the data set. But, mappings most often show defaults and the user is then led astray. According to this notion, a mapping should not be considered when its overall quality is not good enough. Many imperfect mappings can however be exploited by informing the user of the nature and level of defaults. In this work, we propose to visualize local indices trustworthiness and continuity for that purpose.

Original languageEnglish
Title of host publicationProceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010
Pages111-116
Number of pages6
Publication statusPublished - 2010
Externally publishedYes
Event18th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2010 - Bruges
Duration: 28 Apr 201030 Apr 2010

Other

Other18th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2010
CityBruges
Period28/4/1030/4/10

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Lespinats, S., & Aupetit, M. (2010). Mapping without visualizing local default is nonsense. In Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010 (pp. 111-116)

Mapping without visualizing local default is nonsense. / Lespinats, Sylvain; Aupetit, Michael.

Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010. 2010. p. 111-116.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lespinats, S & Aupetit, M 2010, Mapping without visualizing local default is nonsense. in Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010. pp. 111-116, 18th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2010, Bruges, 28/4/10.
Lespinats S, Aupetit M. Mapping without visualizing local default is nonsense. In Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010. 2010. p. 111-116
Lespinats, Sylvain ; Aupetit, Michael. / Mapping without visualizing local default is nonsense. Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010. 2010. pp. 111-116
@inproceedings{a59df7fd349a4e8fbfa1315084aa9d4b,
title = "Mapping without visualizing local default is nonsense",
abstract = "High-dimensional data sets are often embedded in two-dimensional spaces so as to visualize neighborhood relationships. When the map is effective (i.e. when short distances are preserved) it is a powerful way to help an analyst to understand the data set. But, mappings most often show defaults and the user is then led astray. According to this notion, a mapping should not be considered when its overall quality is not good enough. Many imperfect mappings can however be exploited by informing the user of the nature and level of defaults. In this work, we propose to visualize local indices trustworthiness and continuity for that purpose.",
author = "Sylvain Lespinats and Michael Aupetit",
year = "2010",
language = "English",
isbn = "2930307102",
pages = "111--116",
booktitle = "Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010",

}

TY - GEN

T1 - Mapping without visualizing local default is nonsense

AU - Lespinats, Sylvain

AU - Aupetit, Michael

PY - 2010

Y1 - 2010

N2 - High-dimensional data sets are often embedded in two-dimensional spaces so as to visualize neighborhood relationships. When the map is effective (i.e. when short distances are preserved) it is a powerful way to help an analyst to understand the data set. But, mappings most often show defaults and the user is then led astray. According to this notion, a mapping should not be considered when its overall quality is not good enough. Many imperfect mappings can however be exploited by informing the user of the nature and level of defaults. In this work, we propose to visualize local indices trustworthiness and continuity for that purpose.

AB - High-dimensional data sets are often embedded in two-dimensional spaces so as to visualize neighborhood relationships. When the map is effective (i.e. when short distances are preserved) it is a powerful way to help an analyst to understand the data set. But, mappings most often show defaults and the user is then led astray. According to this notion, a mapping should not be considered when its overall quality is not good enough. Many imperfect mappings can however be exploited by informing the user of the nature and level of defaults. In this work, we propose to visualize local indices trustworthiness and continuity for that purpose.

UR - http://www.scopus.com/inward/record.url?scp=84887009297&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887009297&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84887009297

SN - 2930307102

SN - 9782930307107

SP - 111

EP - 116

BT - Proceedings of the 18th European Symposium on Artificial Neural Networks - Computational Intelligence and Machine Learning, ESANN 2010

ER -