1,721,119 research outputs found
Bayesian inference in forensic science
No full textForensic scientists deal with the evaluation of a link between recovered ma- terial of unknown source found at a crime scene and control material coming from a suspect. The assessment of the value of the scientific evidence is typically performed by means of a likelihood ratio, a well established metric in forensic science. However, the derivation of a likelihood ratio may represent a demanding task with several sources of uncertainty, and this has originated a large debate about what should be the most appropriate way to take charge of such uncer- tainty while presenting expressions of evidential value at trial. In such a context, Bayesian networks represent a powerful tool that can be used to study, develop and implement probabilistic procedures for evaluating the probability value of the scientific evidence in forensic science or of an hypothesis of judicial interest
The value of scientific evidence for forensic multivariate data
No full textMultivariate continuous data are becoming more prevalent in forensic science.\ud
Available databases may present a complex dependence structure, with several variables\ud
and several levels of variation. The assessment of the value of evidence can be performed\ud
by the derivation of a likelihood ratio, a rigorous concept that measures the change produced\ud
by a given item of information in the odds in favour of a proposition as opposed\ud
to another, when going from the prior to the posterior distribution. The derivation of a\ud
likelihood ratio may be a demanding task, essentially because of the complexity of the\ud
scenario at hand and the possible poor information at the forensic examiner’s disposal.\ud
This opened the door in the forensic community to a large debate about what should be\ud
the most appropriate way to take charge of uncertainty while presenting expressions of\ud
evidential value at trial. These ideas will be illustrated with reference to a comparative\ud
handwriting scenario
Probability Calculus, Pitfalls of Intuition and Judicial Errors
No full textIn the last decades the powerful of the laboratories’ analytical results to highlight the dynamic of criminal acts has been object of an increasing attention, partly because of the enormous potentiality of the DNA test that has transformed forensic sciences. Statistics is assuming a fundamental role in forensic science, in particular to quantify the probative value of traces collected at the crime scene, or on individuals suspected to be involved in a crime, and analysed by forensic laboratories. Results of comparative analyses (comparative in the sense that some characteristics of traces recovered at the crime scene are compared to those of material originating from the suspect) are generally presented in numeric form, mostly in probability terms, and need to be carefully interpreted. Moreover, the scientific evidence collected at the
crime scene must be considered in conjunction with the bulk of informations the court possesses about the case. Judicial errors correlated to an improper use and
interpretation of probability calculus have originated a large debate about (1) how the probative value of the traces is computed, (2) how to interpret the statistical
expertise at trial, and (3) how the probative value of the scientific evidence can be combined with the other sources of information (for example, a witness statement)
in view of the court decision. Starting from a general description of the potential scenario of interest (section “The Evaluation of Scientific Evidence”), followed by
a non exhaustive overview (section “Pitfalls of Intuition”) of the possible errors of interpretation that can be originated by intuition which may be a bad substitute of
the laws of probability calculus, this paper shows (section “The Transition: Bayes’ Theorem”) how the Bayes’ theorem can represent a formidable tool to combine in a
rational way the different sources of information that may become available at trial
A Bayesian semiparametric approach to the estimation of the covariance structure of space-time data
No full textLa statistica nei tribunali
No full textThis paper draws attention to the increasing role statistics is playing in the courts. The assessment of scientific evidence is often couched in terms of probabilities as measures of uncertainty. It is crucial that probabilistic calculations are accurate and that common fallacies
are avoided. Bayes’ theorem and Bayesian Networks are fundamental tools of the kind of probability calculus which is of interest to courts in determining complex cases
A Bayesian estimation of a separable spatio-temporal model with autoregressive temporal component
No full textDiscussion on the paper by Cowell, Graversen, Lauritzen and Mortera (Analysis of forensic DNA mixtures with artefacts)
No full textDNA is now routinely used in criminal investigations and court cases, although DNA samples taken at crime scenes are of varying quality and therefore present challenging problems for their interpretation. We present a statistical model for the quantitative peak information obtained from an electropherogram of a forensic DNA sample and illustrate its potential use for the analysis of criminal cases. In contrast with most previously used methods, we directly model the peak height information and incorporate important artefacts that are associated with the production of the electropherogram. Our model has a number of unknown parameters, and we show that these can be estimated by the method of maximum likelihood in the presence of multiple unknown individuals contributing to the sample, and their approximate standard errors calculated; the computations exploit a Bayesian network representation of the model. A case example from a UK trial, as reported in the literature, is used to illustrate the efficacy and use of the model, both in finding likelihood ratios to quantify the strength of evidence, and in the deconvolution of mixtures for finding likely profiles of the individuals contributing to the sample. Our model is readily extended to simultaneous analysis of more than one mixture as illustrated in a case example. We show that the combination of evidence from several samples may give an evidential strength which is close to that of a single-source trace and thus modelling of peak height information provides a potentially very efficient mixture analysis
The decisionalization of individualization
Full text linkThroughout forensic science and adjacent branches, academic researchers and practitioners continue to diverge in their perception and understanding of the notion of ‘individualization’, that is the claim to reduce a pool of potential donors of a forensic trace to a single source. In particular, recent shifts to refer to the practice of individualization as a decision have been revealed as being a mere change of label [1], leaving fundamental changes in thought and understanding still pending. What is more, professional associations and practitioners shy away from embracing the notion of decision in terms of the formal theory of decision in which individualization may be framed, mainly because of difficulties to deal with the measurement of desirability or undesirability of the consequences of decisions (e.g., using utility functions). Building on existing research in the area, this paper presents and discusses fundamental concepts of utilities and losses with particular reference to their application to forensic individualization. The paper emphasizes that a proper appreciation of decision tools not only reduces the number of individual assignments that the application of decision theory requires, but also shows how such assignments can be meaningfully related to constituting features of the real-world decision problem to which the theory is applied. It is argued that the decisonalization of individualization requires such fundamental insight to initiate changes in the fields’ underlying understandings, not merely in their label
Posterior likelihood ratios for evaluation of forensic trace evidence given a two-level model on the data by Alberink et al (2013)
No full textNormative decision analysis in forensic science
No full textThis paper focuses on the normative analysis – in the sense of the classic decision-theoretic formulation – of decision problems that arise in con- nection with forensic expert reporting. We distinguish this analytical account from other common types of decision analyses, such as descriptive approaches. While decision theory is, since several decades, an extensively discussed topic in legal literature, its use in forensic science is more recent, and with an em- phasis on goals such as the analysis of the logical structure of forensic ex- pert conclusions regarding, for example, propositions of common source of evidential and known materials. Typical examples are so-called identification (or, individualization) decisions, especially categorical conclusions according to which fingermarks (or stains of biological nature, handwriting, etc.) come from a particular a person of interest. We will present and compare ways of stating forensic identification decisions in decision-theoretic terms and explain their underlying rationale. In particular, we will emphasize the importance of viewing this analysis as normative in the sense of providing a reflective rather than a prescriptive reference point against which people in charge of forensic identification decisions may compare their otherwise (possibly) intuitive and informal reasoning, before acting. Normative decision analysis in forensic sci- ence thus provides a vector through which current practice can be articulated, scrutinized and rethought
- …
