EUVIMEDEUVIMED entdecken
Crossref

Support errors and metric masking in constrained EDS composition regression

Geonwoo Jeong

Physica Scripta · 06.11.2026 · 101 (44) · 446008

DOI: 10.1088/1402-4896/aebaf9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physica Scripta, online erschienen am 04.11.2026. Verfasst von Geonwoo Jeong. Die Arbeit nennt 19 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

True-support error can decrease while an elemental composition assigns more mass incorrectly. We demonstrate this discrepancy in constrained regression of 5 keV energy-dispersive x-ray spectra, using 76 audited material–platform records representing 69 reference compositions. With identical compositions held out together, Ridge regression gives a pooled R 2 of 0.236, yet its ternary mean absolute error exceeds that of a uniform-on-support baseline. After one true element is removed, adding the highest-scored absent element lowers ternary true-support MAE by 3.04 percentage points while raising full-vector L 1 error by 33.14 points. An exact mass decomposition attributes the discrepancy to false-positive mass excluded from the support score. For fixed raw scores, Euclidean simplex projection makes full L 1 nondecreasing when the allowed support gains only absent elements. With an internally trained support detector, lowering the threshold from 0.5 to 0.3 recovers true elements in 37 records and improves both mean errors. Nine of the 76 records have lower true-support MAE and higher full L 1 .

Zitieren

APA

Jeong, G. (2026). Support errors and metric masking in constrained EDS composition regression. Physica Scripta, 101(44), 446008. https://doi.org/10.1088/1402-4896/aebaf9

Vancouver

Jeong G. Support errors and metric masking in constrained EDS composition regression. Physica Scripta. 2026;101(44):446008. doi:10.1088/1402-4896/aebaf9

BibTeX
@article{Jeong2026,
  author = {Jeong, Geonwoo},
  title = {Support errors and metric masking in constrained EDS composition regression},
  journal = {Physica Scripta},
  year = {2026},
  volume = {101},
  number = {44},
  pages = {446008},
  doi = {10.1088/1402-4896/aebaf9}
}
Kostenlos starten →