On 13 December we will hold the AGM of the section, followed by a section meeting on recent research in forensic statistics. The timings are as follows:
16.00-16.15 AGM
16.15-17.00 Dr Therese Graversen
17.00-17.45 Dr Gail Robertson
17.45-18.00 discussion
On 13 December we will hold the AGM of the section, followed by a section meeting on recent research in forensic statistics. The timings are as follows:
16.00-16.15 AGM
16.15-17.00 Dr Therese Graversen - IT University of Copenhagen
17.00-17.45 Dr Gail Robertson - University of Edinburgh
17.45-18.00 discussion
And the titles and abstracts for the talks are
Dr Therese Graversen: A systematic statistical approach to forensic identification
Abstract:
I will discuss my success with treating forensic identification problems based on mixed samples of DNA as a classical statistical problem. I start from a fully specified statistical model, and all further computations are performed entirely within the model, which gives a transparent and robust analysis. I will also discuss how I have used statistical reasoning in witness statements to justify the validity of the statistical evaluation of complex DNA evidence. My implementation of the statistical framework, DNAmixtures, has been used in several criminal cases in the UK and Denmark.
Dr Gail Robertson: Bayesian networks and chain event graphs as decision making tools in forensic science
Abstract:
Bayes’ theorem and likelihood ratios are used in forensic statistics to compare evidence supporting different propositions put forward during court proceedings. There is widespread interest among forensic scientists in using Bayesian network models to evaluate the extent to which scientific evidence supports hypotheses proposed by the prosecution and defence. Bayesian networks are primarily used to compare support for source-level propositions, e.g. those concerned with determining the source of samples found at crime scenes such as hair, fibres, and DNA. While comparing source-level propositions is useful, propositions which refer to criminal activities (i.e. those concerned with understanding how a sample came to be at the crime scene) are of more interest to courts. Less work has been done on developing probabilistic methods to assess activity-level propositions, hence finding a method of evaluating evidence for these types of propositions would benefit practitioners. Chain event graphs have been suggested as a decision making tool to assess the extent to which evidence supports event timelines proposed by the prosecution and defence, and may be more appropriate for assessing activity-level propositions. In this study we used Bayesian networks and chain event graphs to combine different types of evidence supporting activity-level propositions from a real-world drug trafficking case. We compared the use of Bayesian networks and chain event graphs in evaluating evidence from activity-level propositions associated with the case and developed a framework for these to be used by practitioners with non-statistical backgrounds. We found that chain event graphs were better suited at evaluating evidence from complex event timelines put forward by the prosecution and defence.
Dr Therese Graversen (IT University of Copenhagen) and Dr Gail Robertson (University of Edinburgh)