Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections

Bárbara C. Benato; Alexandre X. Falcão; Alexandru C. Telea

doi:10.1007/978-3-031-66743-5_11

Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections

Bárbara C. Benato^*
, Alexandre X. Falcão
, Alexandru C. Telea

^*Corresponding author for this work

Universidade Estadual de Campinas

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Understanding how data separation (DS), visual separation (VS), and classifier performance (CP) are related to each other is important for applications in both machine learning and information visualization. A recent study showed that, for a specific machine learning pipeline using a given multidimensional projection technique, high DS leads to high VS and next high CP. However, whether such correlations would stay the same (or not) when using other projection techniques was left open. We fill this gap by evaluating ten projection techniques in a pipeline that uses three contrastive learning methods (SimCLR, SupCon, and their combination) to produce latent spaces and next train and test classifiers for five image datasets of real-world application with human intestinal parasites. Our work identifies two classes of projection techniques – one leading to poor VS and next poor CS regardless of the available DS, and the other showing a good DS-VS-CP correlation. We argue that this last group of projections is a useful instrument in classifier engineering tasks.

Original language	English
Title of host publication	Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers
Editors	A. Augusto de Sousa, Thomas Bashford-Rogers, Alexis Paljic, Mounia Ziat, Christophe Hurter, Helen Purchase, Petia Radeva, Giovanni Maria Farinella, Kadi Bouatouch
Publisher	Springer
Pages	229-255
Number of pages	27
ISBN (Electronic)	978-3-031-66743-5
ISBN (Print)	978-3-031-66742-8
DOIs	https://doi.org/10.1007/978-3-031-66743-5_11
Publication status	Published - 22 Aug 2024
Event	18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2023 - Lisbon, Portugal Duration: 19 Feb 2023 → 21 Feb 2023

Publication series

Name	Communications in Computer and Information Science
Volume	2103 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2023
Country/Territory	Portugal
City	Lisbon
Period	19/02/23 → 21/02/23

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

Keywords

Contrastive learning
Data separation
Dimensionality reduction algorithms
Embedded pseudolabeling
Image classification
Semi-supervised learning
Visual separation

Access to Document

10.1007/978-3-031-66743-5_11

978-3-031-66743-5_11Final published version, 4.71 MBLicence: Taverne

Cite this

Benato, B. C., Falcão, A. X., & Telea, A. C. (2024). Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections. In A. A. de Sousa, T. Bashford-Rogers, A. Paljic, M. Ziat, C. Hurter, H. Purchase, P. Radeva, G. M. Farinella, & K. Bouatouch (Eds.), Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers (pp. 229-255). (Communications in Computer and Information Science; Vol. 2103 CCIS). Springer. https://doi.org/10.1007/978-3-031-66743-5_11

Benato, Bárbara C. ; Falcão, Alexandre X. ; Telea, Alexandru C. / Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections. Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers. editor / A. Augusto de Sousa ; Thomas Bashford-Rogers ; Alexis Paljic ; Mounia Ziat ; Christophe Hurter ; Helen Purchase ; Petia Radeva ; Giovanni Maria Farinella ; Kadi Bouatouch. Springer, 2024. pp. 229-255 (Communications in Computer and Information Science).

@inproceedings{041e8716da4042719c22b46589354d40,

title = "Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections",

abstract = "Understanding how data separation (DS), visual separation (VS), and classifier performance (CP) are related to each other is important for applications in both machine learning and information visualization. A recent study showed that, for a specific machine learning pipeline using a given multidimensional projection technique, high DS leads to high VS and next high CP. However, whether such correlations would stay the same (or not) when using other projection techniques was left open. We fill this gap by evaluating ten projection techniques in a pipeline that uses three contrastive learning methods (SimCLR, SupCon, and their combination) to produce latent spaces and next train and test classifiers for five image datasets of real-world application with human intestinal parasites. Our work identifies two classes of projection techniques – one leading to poor VS and next poor CS regardless of the available DS, and the other showing a good DS-VS-CP correlation. We argue that this last group of projections is a useful instrument in classifier engineering tasks.",

keywords = "Contrastive learning, Data separation, Dimensionality reduction algorithms, Embedded pseudolabeling, Image classification, Semi-supervised learning, Visual separation",

author = "Benato, \{B{\'a}rbara C.\} and Falc{\~a}o, \{Alexandre X.\} and Telea, \{Alexandru C.\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2023 ; Conference date: 19-02-2023 Through 21-02-2023",

year = "2024",

month = aug,

day = "22",

doi = "10.1007/978-3-031-66743-5\_11",

language = "English",

isbn = "978-3-031-66742-8",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "229--255",

editor = "\{de Sousa\}, \{A. Augusto\} and Thomas Bashford-Rogers and Alexis Paljic and Mounia Ziat and Christophe Hurter and Helen Purchase and Petia Radeva and Farinella, \{Giovanni Maria\} and Kadi Bouatouch",

booktitle = "Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers",

}

Benato, BC, Falcão, AX & Telea, AC 2024, Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections. in AA de Sousa, T Bashford-Rogers, A Paljic, M Ziat, C Hurter, H Purchase, P Radeva, GM Farinella & K Bouatouch (eds), Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers. Communications in Computer and Information Science, vol. 2103 CCIS, Springer, pp. 229-255, 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2023, Lisbon, Portugal, 19/02/23. https://doi.org/10.1007/978-3-031-66743-5_11

Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections. / Benato, Bárbara C.; Falcão, Alexandre X.; Telea, Alexandru C.
Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers. ed. / A. Augusto de Sousa; Thomas Bashford-Rogers; Alexis Paljic; Mounia Ziat; Christophe Hurter; Helen Purchase; Petia Radeva; Giovanni Maria Farinella; Kadi Bouatouch. Springer, 2024. p. 229-255 (Communications in Computer and Information Science; Vol. 2103 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections

AU - Benato, Bárbara C.

AU - Falcão, Alexandre X.

AU - Telea, Alexandru C.

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

PY - 2024/8/22

Y1 - 2024/8/22

N2 - Understanding how data separation (DS), visual separation (VS), and classifier performance (CP) are related to each other is important for applications in both machine learning and information visualization. A recent study showed that, for a specific machine learning pipeline using a given multidimensional projection technique, high DS leads to high VS and next high CP. However, whether such correlations would stay the same (or not) when using other projection techniques was left open. We fill this gap by evaluating ten projection techniques in a pipeline that uses three contrastive learning methods (SimCLR, SupCon, and their combination) to produce latent spaces and next train and test classifiers for five image datasets of real-world application with human intestinal parasites. Our work identifies two classes of projection techniques – one leading to poor VS and next poor CS regardless of the available DS, and the other showing a good DS-VS-CP correlation. We argue that this last group of projections is a useful instrument in classifier engineering tasks.

AB - Understanding how data separation (DS), visual separation (VS), and classifier performance (CP) are related to each other is important for applications in both machine learning and information visualization. A recent study showed that, for a specific machine learning pipeline using a given multidimensional projection technique, high DS leads to high VS and next high CP. However, whether such correlations would stay the same (or not) when using other projection techniques was left open. We fill this gap by evaluating ten projection techniques in a pipeline that uses three contrastive learning methods (SimCLR, SupCon, and their combination) to produce latent spaces and next train and test classifiers for five image datasets of real-world application with human intestinal parasites. Our work identifies two classes of projection techniques – one leading to poor VS and next poor CS regardless of the available DS, and the other showing a good DS-VS-CP correlation. We argue that this last group of projections is a useful instrument in classifier engineering tasks.

KW - Contrastive learning

KW - Data separation

KW - Dimensionality reduction algorithms

KW - Embedded pseudolabeling

KW - Image classification

KW - Semi-supervised learning

KW - Visual separation

UR - https://www.scopus.com/pages/publications/85202594236

U2 - 10.1007/978-3-031-66743-5_11

DO - 10.1007/978-3-031-66743-5_11

M3 - Conference contribution

AN - SCOPUS:85202594236

SN - 978-3-031-66742-8

T3 - Communications in Computer and Information Science

SP - 229

EP - 255

BT - Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers

A2 - de Sousa, A. Augusto

A2 - Bashford-Rogers, Thomas

A2 - Paljic, Alexis

A2 - Ziat, Mounia

A2 - Hurter, Christophe

A2 - Purchase, Helen

A2 - Radeva, Petia

A2 - Farinella, Giovanni Maria

A2 - Bouatouch, Kadi

PB - Springer

T2 - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2023

Y2 - 19 February 2023 through 21 February 2023

ER -

Benato BC, Falcão AX, Telea AC. Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections. In de Sousa AA, Bashford-Rogers T, Paljic A, Ziat M, Hurter C, Purchase H, Radeva P, Farinella GM, Bouatouch K, editors, Computer Vision, Imaging and Computer Graphics Theory and Applications - 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics, VISIGRAPP 2023, Revised Selected Papers. Springer. 2024. p. 229-255. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-66743-5_11

Linking Data Separation, Visual Separation, Classifier Performance Using Multidimensional Projections

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this