lnu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
DeforestVis: Behavior Analysis of Machine Learning Models with Surrogate Decision Stumps
Northwestern University, USA.ORCID iD: 0000-0002-9079-2376
Linnaeus University, Faculty of Technology, Department of computer science and media technology (CM). (ISOVIS)ORCID iD: 0000-0002-2901-935X
Utrecht University, Netherlands.
Linnaeus University, Faculty of Technology, Department of computer science and media technology (CM). Linköping University, Sweden. (ISOVIS;DISA)ORCID iD: 0000-0002-0519-2537
2024 (English)In: Computer graphics forum (Print), ISSN 0167-7055, E-ISSN 1467-8659, Vol. 43, no 6, article id e15004Article in journal (Refereed) Published
Abstract [en]

As the complexity of Machine Learning (ML) models increases and their application in different (and critical) domains grows, there is a strong demand for more interpretable and trustworthy ML. A direct, model-agnostic, way to interpret such models is to train surrogate models—such as rule sets and decision trees—that sufficiently approximate the original ones while being simpler and easier-to-explain. Yet, rule sets can become very lengthy, with many if-else statements, and decision tree depth grows rapidly when accurately emulating complex ML models. In such cases, both approaches can fail to meet their core goal—providing users with model interpretability. To tackle this, we propose DeforestVis, a visual analytics tool that offers summarization of the behavior of complex ML models by providing surrogate decision stumps (one-level decision trees) generated with the Adaptive Boosting (AdaBoost) technique. DeforestVis helps users to explore the complexity vs fidelity trade-off by incrementally generating more stumps, creating attribute-based explanations with weighted stumps to justify decision making, and analyzing the impact of rule overriding on training instance allocation between one or more stumps. An independent test set allows users to monitor the effectiveness of manual rule changes and form hypotheses based on case-by-case analyses. We show the applicability and usefulness of DeforestVis with two use cases and expert interviews with data analysts and model developers.

Place, publisher, year, edition, pages
John Wiley & Sons, 2024. Vol. 43, no 6, article id e15004
Keywords [en]
Surrogate model, model understanding, adaptive boosting, machine learning, visual analytics
National Category
Computer Sciences
Research subject
Computer Science, Information and software visualization
Identifiers
URN: urn:nbn:se:lnu:diva-127909DOI: 10.1111/cgf.15004ISI: 001174196500001Scopus ID: 2-s2.0-85185930256OAI: oai:DiVA.org:lnu-127909DiVA, id: diva2:1839375
Funder
ELLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsAvailable from: 2024-02-20 Created: 2024-02-20 Last updated: 2024-10-17Bibliographically approved

Open Access in DiVA

fulltext(2679 kB)9 downloads
File information
File name FULLTEXT01.pdfFile size 2679 kBChecksum SHA-512
f562931df6c350944dd6f206f32843e20541e19b358558cf7a84017459e362d4067543b2b5e50246e2f0fa8a52a01b0f63905353e21827e625b11257a774f770
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Authority records

Chatzimparmpas, AngelosMartins, Rafael MessiasKerren, Andreas

Search in DiVA

By author/editor
Chatzimparmpas, AngelosMartins, Rafael MessiasKerren, Andreas
By organisation
Department of computer science and media technology (CM)
In the same journal
Computer graphics forum (Print)
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 9 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 153 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf