UCL JILL DANDO INSTITUTE OF SECURITY AND CRIME SCIENCE

Jill Dando Institute of Crime Science

• Home Page
• About Us
• People
• Postgraduate Taught Programmes
• Short Courses
• Research
• PhD Training - UCL SECReT
• JDI Centre for the Forensic Sciences
• Events
• Contact Us

Incorporating population structure into forensic Bayesian networks

Bayesian networks are gaining popularity as a graphical tool to communicate the complex probabilistic reasoning required in the evaluation of DNA evidence. Incorporating allelic dependencies that result from population structure within these networks is a relatively new endeavour and this study provides some initial thoughts on how to approach the construction of these networks.We introduce object-oriented Bayesian networks designed to model forensic identification cases while accounting for population structure. Exact and approximate methods are explored including a blocking Gibbs approach, via HUGIN’s application programming interface (API), to model the unknown subpopulation frequencies. We explore forensic paternity examples, including complex cases with missing data. Accounting for population structure within the Bayesian network framework is an important step forward, and illustrates the flexibility this technology provides as a formal tool for handling complex forensic calculations.

Speaker