Bayesian Networks for Probabilistic Inference and Decision Analysis in Forensic Science, 2nd Edition

Chapter 5Evaluation given activity level propositions

This chapter focusses on general methodological aspects of Bayesian network construction for forensic inference based on examination results regarding the so-called transferable materials such as DNA, glass, fibres, paint or gunshot residues. The starting points are general likelihood ratio formulae proposed in scientific literature to deal with the assessment of such findings. The main features of the various Bayesian networks will be illustrated through examples where the propositions of interest refer to an activity (i.e. so-called activity level propositions). The findings will be either expressed in terms of reported corresponding features or described in terms of intrinsic and extrinsic features. The latter level of description will present a further instance of the use of the notation y and x introduced earlier in Section b01-bib-00394.1.

In many examples, it is reasonable to assume that particular material has a direct relationship with a given individual (which may be the suspect). For example, assume a direct primary transfer.¹ For example, DNA found at a crime scene provides a direct connection between the crime scene and the source of the DNA whoever that source is. The reason for this is that biological material is intimately related to each individual. This may not be the case, however, with certain categories ...