Home > Research > Publications & Outputs > Mixed effects modelling for glass category esti...
View graph of relations

Mixed effects modelling for glass category estimation from glass refractive indicies

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Mixed effects modelling for glass category estimation from glass refractive indicies. / Lucy, David; Zadora, Grzegorz.
In: Forensic Science International, Vol. 212, No. 1-3, 10.10.2011, p. 189-197.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Lucy D, Zadora G. Mixed effects modelling for glass category estimation from glass refractive indicies. Forensic Science International. 2011 Oct 10;212(1-3):189-197. doi: 10.1016/j.forsciint.2011.05.030

Author

Lucy, David ; Zadora, Grzegorz. / Mixed effects modelling for glass category estimation from glass refractive indicies. In: Forensic Science International. 2011 ; Vol. 212, No. 1-3. pp. 189-197.

Bibtex

@article{34d70688aa5d47d28d4e0667bab641ad,
title = "Mixed effects modelling for glass category estimation from glass refractive indicies",
abstract = "520 glass fragments were taken from 105 glass items. Each item was either a container, a window, or glass from an automobile. Each of these three classes of use are defined as glass categories. Refractive indexes were measured both before, and after a programme of re-annealing. Because the refractive index of each fragment could not in itself be observed before and after re-annealing, a model based approach was used to estimate the change in refractive index for each glass category. It was found that less complex estimation methods would be equivalent to the full model, and were subsequently used. The change in refractive index was then used to calculate a measure of the evidential value for each item belonging to each glass category. The distributions of refractive index change were considered for each glass category, and it was found that, possibly due to small samples, members of the normal family would not adequately model the refractive index changes within two of the use types considered here. Two alternative approaches to modelling the change in refractive index were used, one employed more established kernel density estimates, the other a newer approach called log-concave estimation. Either method when applied to the change in refractive index was found to give good estimates of glass category, however, on all performance metrics kernel density estimates were found to be slightly better than log-concave estimates, although the estimates from log-concave estimation prossessed properties which had some qualitative appeal not encapsulated in the selected measures of performance. These results and implications of these two methods of estimating probability densities for glass refractive indexes are discussed.",
keywords = "Evidence evaluation, Likelihood ratio , Re-annealing refractive index , Glass, Glass category , MCMC, Kernel density, Log-concave, ECE, Mixture model",
author = "David Lucy and Grzegorz Zadora",
year = "2011",
month = oct,
day = "10",
doi = "10.1016/j.forsciint.2011.05.030",
language = "English",
volume = "212",
pages = "189--197",
journal = "Forensic Science International",
issn = "1872-6283",
publisher = "Elsevier Ireland Ltd",
number = "1-3",

}

RIS

TY - JOUR

T1 - Mixed effects modelling for glass category estimation from glass refractive indicies

AU - Lucy, David

AU - Zadora, Grzegorz

PY - 2011/10/10

Y1 - 2011/10/10

N2 - 520 glass fragments were taken from 105 glass items. Each item was either a container, a window, or glass from an automobile. Each of these three classes of use are defined as glass categories. Refractive indexes were measured both before, and after a programme of re-annealing. Because the refractive index of each fragment could not in itself be observed before and after re-annealing, a model based approach was used to estimate the change in refractive index for each glass category. It was found that less complex estimation methods would be equivalent to the full model, and were subsequently used. The change in refractive index was then used to calculate a measure of the evidential value for each item belonging to each glass category. The distributions of refractive index change were considered for each glass category, and it was found that, possibly due to small samples, members of the normal family would not adequately model the refractive index changes within two of the use types considered here. Two alternative approaches to modelling the change in refractive index were used, one employed more established kernel density estimates, the other a newer approach called log-concave estimation. Either method when applied to the change in refractive index was found to give good estimates of glass category, however, on all performance metrics kernel density estimates were found to be slightly better than log-concave estimates, although the estimates from log-concave estimation prossessed properties which had some qualitative appeal not encapsulated in the selected measures of performance. These results and implications of these two methods of estimating probability densities for glass refractive indexes are discussed.

AB - 520 glass fragments were taken from 105 glass items. Each item was either a container, a window, or glass from an automobile. Each of these three classes of use are defined as glass categories. Refractive indexes were measured both before, and after a programme of re-annealing. Because the refractive index of each fragment could not in itself be observed before and after re-annealing, a model based approach was used to estimate the change in refractive index for each glass category. It was found that less complex estimation methods would be equivalent to the full model, and were subsequently used. The change in refractive index was then used to calculate a measure of the evidential value for each item belonging to each glass category. The distributions of refractive index change were considered for each glass category, and it was found that, possibly due to small samples, members of the normal family would not adequately model the refractive index changes within two of the use types considered here. Two alternative approaches to modelling the change in refractive index were used, one employed more established kernel density estimates, the other a newer approach called log-concave estimation. Either method when applied to the change in refractive index was found to give good estimates of glass category, however, on all performance metrics kernel density estimates were found to be slightly better than log-concave estimates, although the estimates from log-concave estimation prossessed properties which had some qualitative appeal not encapsulated in the selected measures of performance. These results and implications of these two methods of estimating probability densities for glass refractive indexes are discussed.

KW - Evidence evaluation

KW - Likelihood ratio

KW - Re-annealing refractive index

KW - Glass

KW - Glass category

KW - MCMC

KW - Kernel density

KW - Log-concave

KW - ECE

KW - Mixture model

U2 - 10.1016/j.forsciint.2011.05.030

DO - 10.1016/j.forsciint.2011.05.030

M3 - Journal article

VL - 212

SP - 189

EP - 197

JO - Forensic Science International

JF - Forensic Science International

SN - 1872-6283

IS - 1-3

ER -