Search

Recent advances in complex automated handwriting identification systems have led to a lack of understandability of these systems' computational processes and features by the forensic handwriting examiners that they are designed to support. To mitigate this issue, this research studied the relationship between two systems: FLASH ID® , an automated handwriting/black box system that uses measurements extracted from a static image of handwriting, and MovAlyzeR® , a system that captures kinematic features from pen strokes. For this study, 33 writers each wrote 60 phrases from the London Letter using cursive writing and handprinting, which led to thousands of sample pairs for analysis. The dissimilarities between pairs of samples were calculated using two score functions (one for each system). The observed results indicate that dissimilarity scores based on kinematic spatial-geometric pen stroke features (e.g., amplitude and slant) have a statistically significant relationship with dissimilarity scores obtained using static, graph-based features used by the FLASH ID® system. Similar relationships were observed for temporal features (e.g., duration and velocity) but not pen pressure, and for both handprinting and cursive samples. These results strongly imply that both the current implementation of FLASH ID® and MovAlyzeR® rely on similar features sets when measuring differences in pairs of handwritten samples. These results suggest that studies of biometric discrimination using MovAlyzeR® , specifically those based on the spatial-geometric feature set, support the validity of biometric matching algorithms based on FLASH ID® output.

Aluminum (Al) powder is commonly encountered in improvised explosive devices (IEDs) as a metallic fuel due to its availability and low cost. Although available commercially in powder form, amateur bomb-makers also produce their own Al powder via simple methods found online. In order to provide investigative leads and forensic intelligence, it is important to evaluate not only the composition of homemade devices, but also to distinguish between the various forms of Al powder they contain. To achieve this goal, a method using automated microscopy in combination with statistical techniques has been demonstrated to have the potential to provide source discrimination and investigative leads in source attribution of Al powders in IEDs. The present research refined this method and investigated 59 industrially and amateurly produced Al powder sources with seven subsamples per source using two traditional linear discriminant analyses (LDA), one with a standard data split for training and testing, and another using leave-one-out cross-validation. Averaging the classification accuracies for the two LDA-based analyses, LDA has the ability to correctly classify 59.26%, 83.35%, and 80.69% of the samples based on their powder source, type, and production method, respectively. This classification accuracy represents a 3407%, 317%, and 61.38% increase in accuracy from random class assignment, respectively. Further, in most instances of incorrect data attribution to a particular source, the subsample has been misidentified with another sample of the same powder type or production method.

This article is aimed at the growing number of statisticians interested in the important problem of interpreting evidence within the forensic identification of source problems. Our purpose is to formalize these forensic problems as statistical model selection problems. We use two different classes of statistics for quantifying the evidential value, the likelihood ratio and Bayes Factor. In forensics, both are commonly called the “likelihood ratio approach” and “the value of evidence” despite using different definitions of probability. In statistics, they are closely related to the traditional likelihood ratio from pattern recognition and the Bayes Factor used in model selection. For two different problem frameworks typical in forensic science, the common source and the specific source problems, we show the Bayes Factor and likelihood ratio are not equivalent, and highlight several interesting links between them. These contributions will help to elucidate the effects of choosing different definitions of probability when addressing the forensic identification of source problems. The broader population of statisticians may find this paper interesting as an introduction to forensic applications and for illuminating the connections between model selection methods from two different paradigms of statistics, particularly in view of the active recent discussions on the connections among Bayesian, Fiducial, Frequentist (BFF) approaches.

A National Research Council report on strengthening forensic science raised concern over the lack of scientific studies supporting the validity of examining and interpreting forensic evidence. However, establishing the foundational validity of subjective methods can be challenging. The present study aimed to establish the scientific validity of expert writership opinions and the two-stage approach to evidence interpretation using measures derived from research on handwriting motor control. Regression-based procedures were used to address two experimental questions: 1) what are the relative contributions of kinematic and pressure features in predicting examiner support for alternate writership propositions when examining pairs of questioned handwriting samples; and 2) to what extent does information about the rarity of the kinematic feature dissimilarity scores improve the accuracy of a predictive model based on dissimilarity alone. Regarding the first question, we identified a multifactor model consisting of feature dissimilarity scores and their population distributions having correlation coefficients (R2) of 0.84 and 0.88 for the same-writer and different-writers propositions, respectively. Temporal features contributed up to 21% to the predictive value of the model, whereas spatial features contributed only 9% and pen pressure contributed up to 17%. When we compared models reflecting a single-stage process (based on feature dissimilarities) of forming opinions with models reflecting a two-stage process (based on feature dissimilarities and rarity) we found that the two-stage models had an average of 15.25% greater predictive value than single-stage models. These findings support the scientific validity of FDE writership determinations and underscore the importance of the two-stage approach for evidence interpretation.

Aluminum (Al) powders are commonly used in improvised explosive devices as metallic fuels, a component of explosive mixtures. These powders can be obtained readily from industrial-scale and consumer products, and produced using unsophisticated "kitchen chemistry" techniques. This research demonstrates the potential of automated particle micromorphometry for comparisons between known source and questioned Al powders recovered from IEDs, as well as for insight into the method of Al powder manufacture. Al powder samples were obtained from legitimate manufacturers, and 56 samples were produced "in-house" from Al-containing spray paints and ball-milled Al foils. Transmitted light microscope images of Al powder particles were acquired using an automated stage with automated z-focus; 17 size and shape parameters were measured for all particles. Approximately 37,000-2,500,000 particles/sample were analyzed using an open-source statistical package with customized code. Dimensionality reduction was required for processing the large datasets: eight of the 17 measured variables were selected based on inspection of the correlation matrix. Data from four subsamples from each of the 56 samples produced using "in-house" methods were analyzed using ANOVA to assess the within- and between-sample variation. High within-sample variation was noted; however, ANOVA and post-hoc Tukey's honestly significant difference (HSD) tests demonstrated that the between-sample variation was substantially larger than the within-sample variation. Each sample could be differentiated from all other samples in the test set. Future experiments will focus on ways to reduce the within-sample variation, and additional statistical and microanalytical methods to classify sources and confidently constrain the method of Al powder manufacture.

The two-stage evaluative process is an established framework utilized by forensic document examiners (FDEs) for reaching a conclusion about the source(s) of handwritten evidence. In the second, or discrimination, stage, the examiner attempts to estimate the rarity of observations in a relevant background population. Unfortunately, control samples from a relevant background population are often unavailable, leaving the FDE to reach this determination based on subjective experience. Automated handwriting feature recognition systems are capable of performing both feature comparison and discrimination, yet these systems have not been subjected to empirical validation studies. In the present study, we repurposed a commercially available automated system to generate empirical distributions for ranking feature dissimilarity scores among pairs of handwritten phrases. The blinded results of this automated process were used to survey an international cohort of 36 FDEs regarding their strength of support for same- and different-writer propositions. The survey served to cross-validate FDE decision-making under the two-stage approach. Results from the survey demonstrated a clear pattern of response consistent with ground truth. Predictive regression analyses indicated that the automated feature dissimilarity scores and the log of their cumulative distribution functions accounted for 72% of the variability in FDE opinions. This study demonstrated that feature dissimilarity scores acquired using automated processes and their distributions are closely aligned with FDE decision-making processes supporting the heuristic value of the two-stage evaluative framework.

Recent developments in forensic science have lead to a proliferation of methods for quantifying the probative value of evidence by constructing a Bayes Factor that allows a decision-maker to select...

In the various forensic science disciplines, recent analytical developments paired with modern statistical computational tools have led to the proliferation of adhoc techniques for quantifying the probative value of forensic evidence. Many legal and scientific scholars agree that the value of evidence should be reported as a likelihood ratio or a Bayes Factor. Quantifying the probative value of forensic evidence is subjected to many sources of variability and uncertainty. There is currently a debate on how to characterize the reliability of the value of evidence. Some authors have proposed associating a confidence/credible interval with the value of evidence assigned to a collection of forensic evidence. In this paper, we will discuss the reasons for our opinion that interval quantifications for the value of evidence should not be used directly in the Bayesian decision-making process to determine the support of the evidence for one of the two competing hypotheses.