Essentials of Bayesian networks in forensic science

Essentials of DNA interpretation

Essentials of forensic interpretation

Statistics and the Evaluation of Forensic Evidence

Target audience

Forensic science practitioners, chief forensic scientists or coordinators, including reporting officers, forensic geneticians, lawyers, risk management analysts, having a University degree (at least BSc) or an equivalent degree of a higher education programme

The course is given in English. However, exercises, discussions and one-to-one tutorial can be led in English, German or French.

Introduction

Bayesian networks (BN) are innovative and are gaining more and more widespread use within many professional branches. This on-line course gives a comprehensive view of the use of BNs for the probabilistic evaluation of scientific evidence in forensic science applications. It provides students with room for an incremental and in-depth study of the various aspects that a solid mastery of BNs in practical applications demands. The course content is based on the use of likelihood ratios, an approach that is supported by the International Society of Forensic Genetics, the European Network of Forensic Science Institutes and the Association of Forensic Science Providers. It lasts 6 months with a workload per week of 3 hours on the online platform and 1.5 to 2 hours of personal work (in total around 150 hours training).

The course team also offers further online short courses for practitioners who wish to specialize in DNA interpretation or Forensic interpretation .
And, in addition to these two short courses, a Certificate of Advanced Studies (CAS) in Statistics and the Evaluation of Forensic Evidence is proposed to forensic practioners willing to acquire a comprehensive training (around 470 hours training).

Objectives

• To acquire solid knowledge and understanding of the principles of BNs.
• To master methodology in graphical inference modelling using specialised BN software.
• To develop and apply standard networks that can be implemented in current software systems, and that form the core models which participants can transfer to their analysis of real cases from their own professional environment.

Program

• Introduction to evaluation of scientific evidence and Bayesian networks (BNs)
  • Uncertainty and probability, principles of probabilistic scientific evidence interpretation, relevant propositions, the logic of BNs
• Evaluation using BNs: general issues
  • Issues in one-trace transfer cases (match probabilities, relevance, error rates, multiple propositions), evidence with more than one component, scenarios with more than one stain, software functionalities
• BNs for selected types of scientific evidence
  • DNA, fingermarks, shoe- and toolmarks, transfer evidence (glass, fibres, gunshot residues), drugs and arson
• Applications (case studies)
  • Application of BNs for case based probabilistic reasoning and scientific evidence assessment
• Combination of evidence
  • General issues and case studies
• Further issues
  • Database searching, sampling, introduction to decision theory (Bayesian decision networks), object-orientation

Instructors

• Professor Franco Taroni
• Professor Christophe Champod
• Professor Alex Biedermann

The instructors have theoretical and practical experience with evaluation and interpretation from laboratory to courtroom. They have published over the years numerous scholarly papers and textbooks on the subjects of evaluation and statistics in forensic science ( Taroni et al., «Bayesian networks and Probabilistic Inference in Forensic Science», John Wiley & Sons, 2006 ).

Consultants

• Prof. Aitken, University of Edinburgh, Scotland
• Dr Evett, United Kingdom

Programme coordination

• For academic questions : sefe [a] unil[.]ch

Practical information

Course venue

Online course

Registration

Apply online. A current curriculum vitae is to be uploaded through our online application system.

Course fee

3’500.- Swiss Francs

Registration deadline *

May 17th, 2019


* Beyond the deadline, registrations will be considered depending on availability.

Contact for administrative questions

Ms. Mary-Claire André Mollet
mary-claire.andremollet [a] unil[.]ch

More information »