Forensic applications of Artificial Intelligence
Key words: forensic research, argumentation- and narrative-based communication, cognitive modeling, Bayesian networks, decision support systems.
Themes: Knowledge Technology, Forensic Science, AI and Law
Recent miscarriages of justice have increased the interest from legal practice in scientifically founded ways of treating evidence. Forensic statistics can provide such foundations. However, because of the communication gap between forensic statisticians, crime investigators and lawyers, statistical evidence is easily misinterpreted in court, resulting in wrong decisions. Therefore, methods must be developed to support the communication between the parties involved. Since lawyers are more used to thinking in terms of arguments and scenarios, we propose to develop methods that support argumentation- and narrative-based communication about statistical evidence, building on AI models of argumentation and scenario construction.
We focus on Bayesian Networks, since their graphical structure can be used to express scenarios, while they also support probabilistic reasoning. To draw inferences from Bayesian Networks, lawyers must understand how the evidence was modelled and what the model means. Therefore, support tools will be developed both for the modelling of evidence as a Bayesian Network and for the understanding of the resulting network. To support the modelling of evidence in Bayesian Networks, argumentation tools will be developed to model inferences and disagreement between experts, while narrative tools will be developed to support the construction of alternative scenarios. To aid the understanding of a Bayesian Network, tools will be developed for automatically extracting both arguments and scenarios from a network and for comparing alternative scenarios. The mathematical and computational tools developed in the project will be practically assessed by means of realistic case studies and training sessions in collaboration with forensic legal practice.
The expected end result of these modules will be the production of analytic tools and software for preventing judicial errors and practical tools for supporting legal practitioners. The models will also enable professional exchange between statistical experiments and legal professionals.
Participating researchers: 4
Research programme: Multi-agent Systems
Research Institute: Bernoulli Institute
Faculty: Faculty of Science and Engineering
Graduate school: Graduate School of Science and Engineering (GSSE)
Collaboration: Prof. dr. mr. H. Prakken (Faculty of Law, University of Groningen, Department of Information and Computing Sciences, Utrecht University), Dr. S. Renooij , Prof. dr. J.-J. Ch. Meyer , Dr. F.J. Bex , M. G.C. Haverkate, Mr. A. Bood, Mr. drs. J. Moors, Mr. H.W.G. Stikkelbroeck, B.G.L. Stinissen, Mr. E.J. Willekers.
|Last modified:||03 March 2020 2.51 p.m.|