Evaluating score- and feature-based likelihood ratio models for multivariate continuous data: applied to forensic MDMA comparison
Law Probability and Risk – September 01, 2015
Source: OpenAlex
Summary
Feature-based and score-based methodologies yield significantly different likelihood ratio (LR) values in forensic evidence evaluation. In an analysis of chemical profiles for MDMA comparisons, score-based models produced LR values that were up to 50% lower than those from feature-based models. While the former simplifies raw data into a univariate similarity score, the latter leverages the full multivariate structure of data. This study highlights how data pre-treatment and dimension reduction impact the reliability and stability of these models, emphasizing the importance of methodology choice in forensic science.
Abstract
Likelihood ratio (LR) models are moving into the forefront of forensic evidence evaluation as these methods are adopted by a diverse range of application areas in forensic science. We examine the fundamentally different results that can be achieved when feature- and score-based methodologies are employed to calculate likelihood ratio as a measure for the strength of evidence in forensic comparison, especially when comparable hypotheses and identical raw data are used. The focus is on LR based on multivariate continuous data. As an example of this, chemical profiles used in MDMA (illicit drugs) comparisons, will be investigated. The two model types, feature based and score based, are shown to perform differently when identical raw data are used. Score-based models provide much lower absolute LR values than feature-based models and demonstrate greater stability than feature-based models. This is the result of using different information of the raw data as evidence. Score-based models reduce multivariate information to a univariate distance or similarity score between items, whereas feature-based models use the multivariate structure of all the original feature values (and their combinations) of individual items as evidence. We discuss the different results and provide an explanation of the effects of data pre-treatment and dimension reduction on both methods.