Cross entropy and log likelihood ratio cost as performance measures for multi‐conclusion categorical outcomes scales
Journal of Forensic Sciences · 2025-03 · 70 (2) · 589-606
DOI: 10.1111/1556-4029.15686 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Forensic Sciences, online erschienen am 10.12.2024. Verfasst von Eric M Warren und 2 weiteren Autor:innen. Beteiligte Einrichtung: SEP Forensic Consultants, Memphis, Tennessee, USA. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 94 Literaturangaben.
Abstract (Originalsprache)
The inconclusive category in forensics reporting is the appropriate response in many cases, but it poses challenges in estimating an “error rate”. We discuss the use of a class of information‐theoretic measures related to cross entropy as an alternative set of metrics that allows for performance evaluation of results presented using multi‐category reporting scales. This paper shows how this class of performance metrics, and in particular the log likelihood ratio cost, which is already in use with likelihood ratio forensic reporting methods and in machine learning communities, can be readily adapted for use with the widely used multiple category conclusions scales. Bayesian credible intervals on these metrics can be estimated using numerical methods. The application of these metrics to published test results is shown. It is demonstrated, using these test results, that reducing the number of categories used in a proficiency test from five or six to three increases the cross entropy, indicating that the higher number of categories was justified, as it they increased the level of agreement with ground truth.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- SEP Forensic Consultants, Memphis, Tennessee, USA.
- Simon Business School, University of Rochester, Rochester, New York, USA.
- Computer and Data Sciences, Merrimack College, North Andover, Massachusetts, USA.
Zitieren
APA
Warren, E. M., Handley, J. C., & Sheets, H. D. (2025). Cross entropy and log likelihood ratio cost as performance measures for multi‐conclusion categorical outcomes scales. Journal of Forensic Sciences, 70(2), 589-606. https://doi.org/10.1111/1556-4029.15686
Vancouver
Warren EM, Handley JC, Sheets HD. Cross entropy and log likelihood ratio cost as performance measures for multi‐conclusion categorical outcomes scales. Journal of Forensic Sciences. 2025;70(2):589-606. doi:10.1111/1556-4029.15686
BibTeX
@article{Warren2025,
author = {Warren, Eric M and Handley, John C and Sheets, H David},
title = {Cross entropy and log likelihood ratio cost as performance measures for multi‐conclusion categorical outcomes scales},
journal = {Journal of Forensic Sciences},
year = {2025},
volume = {70},
number = {2},
pages = {589--606},
doi = {10.1111/1556-4029.15686}
}
EUVIMED