O'Reilly logo

Bayesian Networks for Probabilistic Inference and Decision Analysis in Forensic Science, 2nd Edition by Colin Aitken, Paolo Garbolino, Silvia Bozza, Alex Biedermann, Franco Taroni

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Chapter 4Evaluation given source level propositions

4.1 General considerations

Evaluation of scientific results in the light of source level propositions typically involves a rather restricted number of propositions. On the one hand, the scientist needs to express the outcomes of a comparison between material found on a crime scene or on a person of interest and control material from a potential source. Depending on the desired level of detail, this may be achieved using one or more propositions. On the other hand, the scientist needs to formulate competing propositions that state whether or not a given potential source is the true source. Throughout forensic literature, there is considerable variation in the notation used in formal analyses involving source level propositions, whereas, on a conceptual account, there is broad agreement with respect to the relevance relationships assumed between the various variables.

Bayesian networks are well suited to clarify such distinctions. Figure 4.1 exemplifies three recurrent expressions that are encountered at different junctures throughout this book. Start by considering Figure 4.1(i), which represents Bayes' theorem in one of its most general forms. Here, the variable H represents the source level propositions c04-math-0002 and c04-math-0003, whereas E is an ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required