Dimensionality reduction of multielement glass evidence to calculate likelihood ratios

Gupta, A; Corzo, R; Akmeemana, A; Lambert, K; Jimenez, K; Curran, JM; Almirall, JR

Dimensionality reduction of multielement glass evidence to calculate likelihood ratios

Gupta, A

Corzo, R Akmeemana, A Lambert, K Jimenez, K Curran, JM Almirall, JR

Permalink

Publisher:: WILEY
Publication Type:: Journal Article
Citation:: Journal of Chemometrics, 2020, 35, (1)
Issue Date:: 2020-01-01

Closed Access

	Filename	Description	Size
	PublishedInIssue.pdf	Published version	8.84 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Gupta, A https://orcid.org/0000-0003-4957-4008
dc.contributor.author	Corzo, R
dc.contributor.author	Akmeemana, A
dc.contributor.author	Lambert, K
dc.contributor.author	Jimenez, K
dc.contributor.author	Curran, JM
dc.contributor.author	Almirall, JR
dc.date.accessioned	2021-02-03T08:17:47Z
dc.date.available	2021-02-03T08:17:47Z
dc.date.issued	2020-01-01
dc.identifier.citation	Journal of Chemometrics, 2020, 35, (1)
dc.identifier.issn	0886-9383
dc.identifier.issn	1099-128X
dc.identifier.uri	http://hdl.handle.net/10453/145769
dc.description.abstract	© 2020 John Wiley & Sons, Ltd. Dimensionality reduction of multivariate elemental concentrations of glass is reported for computing likelihood ratios (LRs). The LRs calculated using principal component analysis (PCA) and a post hoc calibration step result in very low (<1%) false inclusions when comparing glass samples known to originate from different sources and very low (<1%) false exclusions when comparing glass samples known to originate from the same source. The LRs calculated using the novel PCA approach are compared with previously reported LRs calculated using a more computationally intensive Multivariate Kernel (MVK) model followed by a calibration step using a pool adjacent violators (PAV) algorithm. In both cases, the calibrated LRs limited the magnitude of the misleading evidence, providing only weak to moderate support for the incorrect hypotheses. Most of the different pairs that were found to be falsely included were explained by chemical relatedness (same manufacturer of the glass sources in very close time interval between manufacture). The computation of LRs using dimensionality reduction of elemental concentrations using PCA may transfer to other multivariate data-generating evidence types.
dc.language	English
dc.publisher	WILEY
dc.relation.ispartof	Journal of Chemometrics
dc.relation.isbasedon	10.1002/cem.3298
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.subject	0301 Analytical Chemistry
dc.subject.classification	Analytical Chemistry
dc.title	Dimensionality reduction of multielement glass evidence to calculate likelihood ratios
dc.type	Journal Article
utslib.citation.volume	35
utslib.for	0301 Analytical Chemistry
pubs.organisational-group	/University of Technology Sydney/Faculty of Science
pubs.organisational-group	/University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
dc.date.updated	2021-02-03T08:17:31Z
pubs.issue	1
pubs.publication-status	Published
pubs.volume	35
utslib.citation.issue	1

Abstract:

© 2020 John Wiley & Sons, Ltd. Dimensionality reduction of multivariate elemental concentrations of glass is reported for computing likelihood ratios (LRs). The LRs calculated using principal component analysis (PCA) and a post hoc calibration step result in very low (<1%) false inclusions when comparing glass samples known to originate from different sources and very low (<1%) false exclusions when comparing glass samples known to originate from the same source. The LRs calculated using the novel PCA approach are compared with previously reported LRs calculated using a more computationally intensive Multivariate Kernel (MVK) model followed by a calibration step using a pool adjacent violators (PAV) algorithm. In both cases, the calibrated LRs limited the magnitude of the misleading evidence, providing only weak to moderate support for the incorrect hypotheses. Most of the different pairs that were found to be falsely included were explained by chemical relatedness (same manufacturer of the glass sources in very close time interval between manufacture). The computation of LRs using dimensionality reduction of elemental concentrations using PCA may transfer to other multivariate data-generating evidence types.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/145769