Deconvolution of Dust Mixtures by Latent Dirichlet Allocation in Forensic Science

• NCJ Number: 254153
• Author(s): Madeline Ausdemore; Cedric Neumann
• Date Published: 2019
• Length: 11 pages
• Grant Number(s): 2014-IJ-CX-K088
• Agencies: NIJ-Sponsored
• Publication Type: Research (Applied/Empirical); Report (Study/Research); Report (Grant Sponsored); Program/Project Description
• Award: Foundational Research into the Quantification of the Value of Forensic Evidence for Complex Evidential Forms arising from Impression and Pattern Evidence

Annotation

This article explains a technique for analyzing dust particles on a person's shoe that indicate whether a person has been in the vicinity of a crime scene.

Abstract

Dust particles recovered from the soles of shoes may be indicative of the sites recently visited by an individual and the presence of an individual at a particular site of interest, e.g., the scene of a crime. By describing the dust profile of a given site by a multinomial distribution over a fixed number of dust particle types, investigators can define the probability distribution of the mixture of dust recovered from the sole of a shoe via Latent Dirichlet Allocation. The current study used Variational Bayesian Inference to study the parameters of the model and used their resulting posterior distributions to make inference on (a) the contributions of sites of interest to a dust mixture, and (b) the particle profiles associated with these sites. (publisher abstract modified)