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 12Object-oriented networks

12.1 Object orientation

In some areas of application in forensic science, inference and decision problems may be characterized by recurrent appearances of the same or similar components. Forensic kinship analyses (Section 7.9) provide an illustrative example for this. They typically involve multiple individuals, each being described in terms of their genotype. Graphical models for such applications can reflect this property in terms of repetitive sub-structures, also sometimes called (local) network fragments (for examples, see Figure 7.16). From a methodological point of view, decomposing a problem into smaller parts, in particular by recognizing generic structures and describing these in terms of local structures, can help to assure coherence in the overall model construction process. On a practical account, however, working with local structures can also bring complications. For example, when the analyst judges that a local structure needs definitional changes, then all instances where the structure of interest appears might also need a revision. More generally, the differences in focus on either local structures or aggregations of local structures within a larger context relate to the notion of hierarchy. This notion also reflects the way in which analysts tend to reason about problems. In fact, they commonly shift focus between different levels of abstractions, depending on the requirements of the situation at hand.

In order to support ...

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