BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:Modelling competing Legal Arguments using Bayesian Model Compariso n and Averaging - Martin Neil (Queen Mary University of London) DTSTART:20160928T103000Z DTEND:20160928T111500Z UID:TALK67666@talks.cam.ac.uk CONTACT:INI IT DESCRIPTION:Bayesian models of legal arguments generally aim to prod uce a single integrated model\, combining each of the legal arguments unde r consideration. This combined approach implicitly assumes that variables and their relationships can be represented without any contradiction or mi salignment and in a way that makes sense with respect to the competing arg ument narratives. In contrast to this integrated approach\, Non-Bayesian a pproaches to legal argumentation have tended to be narrative based and hav e focused on comparisons between competing stories and explanations. This paper describes a novel approach to compare and &lsquo\;average&rsquo\; Ba yesian models of legal arguments that have been built independently and wi th no attempt to make them consistent in terms of variables\, causal assum ptions or parameterization. The approach is consistent with subjectivist B ayesian philosophy. Practically\, competing models of legal arguments are assessed by the extent to which the credibility of the sources of evidence are confirmed or disconfirmed in court. Those models that are more heavil y disconfirmed are assigned lower weights\, as model confidence measures\, in the Bayesian model comparison and averaging approach adopted. In this way plurality of arguments are allowed yet a single judgement based on all arguments is possible and rational.

Authors: Prof. Martin Neil (Qu een Mary\, University of London)\, Prof. Norman Fenton (Queen Mary\, Unive rsity of London)\, Prof David Lagnado (UCL)\, and Prof Richard Gill (Leiden University) LOCATION:Seminar Room 1\, Newton Institute END:VEVENT END:VCALENDAR