A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature

Geoffrey Stewart Morrison; Daniel Ramos; Rolf JF Ypma; Nabanita Basu; Kim de Bie; Ewald Enzinger; Zeno Geradts; Didier Meuwly; David van der Vloed; Peter Vergeer; Philip Weber

doi:10.1016/j.fsisyn.2022.100230

A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature

Geoffrey Stewart Morrison^*, Daniel Ramos, Rolf JF Ypma, Nabanita Basu, Kim de Bie, Ewald Enzinger, Zeno Geradts, Didier Meuwly, David van der Vloed, Peter Vergeer, Philip Weber

^*Corresponding author for this work

Research output: Contribution to journal › Letter, comment/opinion or interview › peer-review

Abstract

We agree wholeheartedly with Biedermann (2022) FSI Synergy article 100222 in its criticism of research publications that treat forensic inference in source attribution as an “identification” or “individualization” task. We disagree, however, with its criticism of the use of machine learning for forensic inference. The argument it makes is a strawman argument. There is a growing body of literature on the calculation of well-calibrated likelihood ratios using machine-learning methods and relevant data, and on the validation under casework conditions of such machine-learning-based systems.

Original language	English
Article number	100230
Number of pages	2
Journal	Forensic Science International: Synergy
Volume	4
Early online date	6 May 2022
DOIs	https://doi.org/10.1016/j.fsisyn.2022.100230
Publication status	Published - 19 May 2022

Bibliographical note

© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license 4.0

Funding Information:
The writing of this response was supported by Research England's Expanding Excellence in England Fund as part of funding for the Aston Institute for Forensic Linguistics 2019–2023.

Keywords

Forensic inference
Machine learning

Access to Document

10.1016/j.fsisyn.2022.100230Licence: CC BY 4.0

A response to Biedermann (2022)
© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license 4.0
Final published version, 280 KBLicence: CC BY 4.0

Cite this

Morrison, G. S., Ramos, D., Ypma, R. JF., Basu, N., Bie, K. D., Enzinger, E., Geradts, Z., Meuwly, D., Vloed, D. V. D., Vergeer, P., & Weber, P. (2022). A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature. Forensic Science International: Synergy, 4, Article 100230. https://doi.org/10.1016/j.fsisyn.2022.100230

@article{114b57f93a0f49f99d14f930634fe656,

title = "A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature",

abstract = "We agree wholeheartedly with Biedermann (2022) FSI Synergy article 100222 in its criticism of research publications that treat forensic inference in source attribution as an “identification” or “individualization” task. We disagree, however, with its criticism of the use of machine learning for forensic inference. The argument it makes is a strawman argument. There is a growing body of literature on the calculation of well-calibrated likelihood ratios using machine-learning methods and relevant data, and on the validation under casework conditions of such machine-learning-based systems.",

keywords = "Forensic inference, Machine learning",

author = "Morrison, {Geoffrey Stewart} and Daniel Ramos and Ypma, {Rolf JF} and Nabanita Basu and Bie, {Kim de} and Ewald Enzinger and Zeno Geradts and Didier Meuwly and Vloed, {David van der} and Peter Vergeer and Philip Weber",

note = "{\textcopyright} 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license 4.0 Funding Information: The writing of this response was supported by Research England's Expanding Excellence in England Fund as part of funding for the Aston Institute for Forensic Linguistics 2019–2023. ",

year = "2022",

month = may,

day = "19",

doi = "10.1016/j.fsisyn.2022.100230",

language = "English",

volume = "4",

}

Morrison, GS, Ramos, D, Ypma, RJF, Basu, N, Bie, KD, Enzinger, E, Geradts, Z, Meuwly, D, Vloed, DVD, Vergeer, P & Weber, P 2022, 'A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature', Forensic Science International: Synergy, vol. 4, 100230. https://doi.org/10.1016/j.fsisyn.2022.100230

A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature. / Morrison, Geoffrey Stewart; Ramos, Daniel; Ypma, Rolf JF et al.
In: Forensic Science International: Synergy, Vol. 4, 100230, 19.05.2022.

Research output: Contribution to journal › Letter, comment/opinion or interview › peer-review

TY - JOUR

T1 - A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature

AU - Morrison, Geoffrey Stewart

AU - Ramos, Daniel

AU - Ypma, Rolf JF

AU - Basu, Nabanita

AU - Bie, Kim de

AU - Enzinger, Ewald

AU - Geradts, Zeno

AU - Meuwly, Didier

AU - Vloed, David van der

AU - Vergeer, Peter

AU - Weber, Philip

N1 - © 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license 4.0 Funding Information: The writing of this response was supported by Research England's Expanding Excellence in England Fund as part of funding for the Aston Institute for Forensic Linguistics 2019–2023.

PY - 2022/5/19

Y1 - 2022/5/19

N2 - We agree wholeheartedly with Biedermann (2022) FSI Synergy article 100222 in its criticism of research publications that treat forensic inference in source attribution as an “identification” or “individualization” task. We disagree, however, with its criticism of the use of machine learning for forensic inference. The argument it makes is a strawman argument. There is a growing body of literature on the calculation of well-calibrated likelihood ratios using machine-learning methods and relevant data, and on the validation under casework conditions of such machine-learning-based systems.

AB - We agree wholeheartedly with Biedermann (2022) FSI Synergy article 100222 in its criticism of research publications that treat forensic inference in source attribution as an “identification” or “individualization” task. We disagree, however, with its criticism of the use of machine learning for forensic inference. The argument it makes is a strawman argument. There is a growing body of literature on the calculation of well-calibrated likelihood ratios using machine-learning methods and relevant data, and on the validation under casework conditions of such machine-learning-based systems.

KW - Forensic inference

KW - Machine learning

UR - https://www.sciencedirect.com/science/article/pii/S2589871X22000158?via%3Dihub

U2 - 10.1016/j.fsisyn.2022.100230

DO - 10.1016/j.fsisyn.2022.100230

M3 - Letter, comment/opinion or interview

C2 - 35647509

SN - 2589-871X

VL - 4

JO - Forensic Science International: Synergy

JF - Forensic Science International: Synergy

M1 - 100230

ER -

A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this