Skip to content

Ensemble SLRs for Forensic Evidence Comparison

Primary Author: Danica Ommen
Type: Webinar

This CSAFE webinar was held on August 25, 2022.

Presenter:
Danica Ommen
Assistant Professor – Department of Statistics, Iowa State University

Presentation Description:

To strengthen the statistical foundations of forensic evidence interpretation, likelihood ratios and Bayes factors are advocated to quantify the value of evidence. Both methods rely on formulating a statistical model, which can be challenging for complex evidence. Machine learning score-based likelihood ratios have been proposed as an alternative in those cases. Under this framework, a (dis)similarity score and its distribution under alternative propositions are estimated using pairwise comparisons, but pairwise comparisons of all the evidential objects result in dependent scores. While machine learning methods may not require distributional assumptions, most assume independence. We introduce a sampling and ensembling approach to remedy this lack of independence. We generate sets where assumptions are met to develop multiple score-based  likelihood ratios later aggregated into a final score to quantify the value of evidence.

Webinar Recording:

Read more about the study in this post

Related Resources

Source Camera Identification with Multi-Camera Smartphones

Source Camera Identification with Multi-Camera Smartphones

An overview of source camera identification on multi-camera smartphones, and introduction to the new CSAFE multi-camera smartphone image database, and a summary of recent results on the iPhone 14 Pro’s.
What’s in a Name? Consistency in Latent Print Examiners’ Naming Conventions and Perceptions of Minutiae Frequency

What’s in a Name? Consistency in Latent Print Examiners’ Naming Conventions and Perceptions of Minutiae Frequency

Fingerprint minutia types influence LPEs’ decision-making processes during analysis and evaluation, with features perceived to be rarer generally given more weight. However, no large-scale studies comparing examiner perceptions of minutiae…
An alternative statistical framework for measuring proficiency

An alternative statistical framework for measuring proficiency

Item Response Theory, a class of statistical methods used prominently in educational testing, can be used to measure LPE proficiency in annual tests or research studies, while simultaneously accounting for…
Examiner variability in pattern evidence: proficiency, inconclusive tendency, and reporting styles

Examiner variability in pattern evidence: proficiency, inconclusive tendency, and reporting styles

The current approach to characterizing uncertainty in pattern evidence disciplines has focused on error rate studies, which provide aggregated error rates over many examiners and pieces of evidence. However, decisions…