Hide menu

Probabilistic inference in Forensic Interpretation and related medical decision problems with applications on DNA analysis

2010VT

Status Archive
School Computer and Information Science (CIS)
Division STIMA
Owner Anders Nordgaard

  Log in  




Course plan

No of lectures

Recommended for

PhD students within areas of natural and medical science, who need training in probabilistic reasoning and decision theory.

The course was last given

Goals

The course provides knowledge about probabilistic evaluation of forensic evidence, in particular DNA evidence, and skills for the evaluation of evidence and decision making using graphical models, in particular Bayesian networks.

Having completed the course, the student should be able to:
- use knowledge about
 probability models for the value of evidence,
 population genetics aspects of likelihoods of DNA profiles
- display a good understanding of major principles for
 the construction of a Bayesian network for evidence evaluation
 the inclusion of different types of evidence in a Bayesian network
 the application of Bayesian networks at different stages of a case and with different levels of propositions (source level, activity level and offence level)
 the application of Bayesian networks on decision making within medical and biological science
- use established software to analyse Bayesian networks,
- demonstrate insightful assessment about the capacity of different models (graphical and non-graphical) for evaluation of DNA evidence in particular case-work.

Prerequisites

Student’s entering the course should have passed at least one course in basic probability theory. A basic course in calculus is also required.

Organization

The teaching is limited to individual supervision.

Contents

The course content comprises:
- likelihood ratios for the weight-of-evidence,
- population genetics models for probability calculation,
- identification through DNA profiles,
- relatedness analysis,
- common errors of interpretation,
- graphical models in probability reasoning,
- propositions at different levels,
- Bayesian networks for DNA evidence,
- transfer evidence and the combination of evidence,
- analyses at different stages, pre-assessment
- continuous Bayesian networks
- Bayesian networks as a decision support tool in medical and biological science
- strengths and weaknesses of probability models and Bayesian networks
- established software for Bayesian networks

Literature

Balding, D.J. Weight-of-evidence for Forensic DNA profiles. Chichester: Wiley, 2005.
Taroni, F., Aitken, C, Garbolino, P. & Biedermann, A. Bayesian Networks and Probabilistic Inference in Forensic Science. Chichester: Wiley, 2006.

Lecturers

Examiner

Anders Nordgaard

Examination

Assignments encompassing computer-based data analysis. One final oral examination.

Credit

10

Comments

The course is established to serve the PhD program in forensic genetics at The National Board of Forensic Medicine. It is also established to be given as a pilot course to the forthcoming course "Bayesian networks with forensic and other applications" to be given 2010HT. No costs will be defined for this course.


Page responsible: Director of Graduate Studies