Constructing and Visualizing High-Quality Classifier Decision Boundary Maps dagger

Francisco C. M. Rodrigues*, Mateus Espadoto, Roberto Hirata Jr, Alexandru C. Telea

*Corresponding author voor dit werk

Onderzoeksoutput: ArticleAcademicpeer review

32 Citaten (Scopus)
168 Downloads (Pure)

Samenvatting

Visualizing decision boundaries of machine learning classifiers can help in classifier design, testing and fine-tuning. Decision maps are visualization techniques that overcome the key sparsity-related limitation of scatterplots for this task. To increase the trustworthiness of decision map use, we perform an extensive evaluation considering the dimensionality-reduction (DR) projection techniques underlying decision map construction. We extend the visual accuracy of decision maps by proposing additional techniques to suppress errors caused by projection distortions. Additionally, we propose ways to estimate and visually encode the distance-to-decision-boundary in decision maps, thereby enriching the conveyed information. We demonstrate our improvements and the insights that decision maps convey on several real-world datasets.

Originele taal-2English
Artikelnummer280
Aantal pagina's22
TijdschriftAHF-Information
Volume10
Nummer van het tijdschrift9
DOI's
StatusPublished - sep.-2019

Vingerafdruk

Duik in de onderzoeksthema's van 'Constructing and Visualizing High-Quality Classifier Decision Boundary Maps dagger'. Samen vormen ze een unieke vingerafdruk.

Citeer dit