Atkinson, Katie; Collenette, Joe; Bench-Capon, Trevor; Dzehtsiarou, Kanstantsin
Abstract: One approach to building legal support systems is to build a model of the relevant legal knowledge, put this model into an executable form, and then run this through an interface designed to collect information about particular cases from the user and provide explanations. This approach has its origins in formalising statutes as logic programs, and is still in use today: for example, the ANGELIC methodology models the domain as an Abstract Dialectical Framework. The usability of such systems depends on the terms used in the law being represented: the system may arrive at questions which users can answer, but often this is not the case and only users familiar with the practice and application of the law will be able to supply the required information. Earlier work applied this approach to the European Convention on Human Rights (ECHR). Although the performance of the tool built for that domain was good, the questions posed to the user demanded a good deal of knowledge and experience of the ECHR. In this paper we drive deeper down into one important aspect of ECHR decisions, namely the admissibility of claims. We use the knowledge of an expert with extensive experience of the ECHR to extend the ADF, through intermediate levels, to identify questions that are appropriate to the target user. We further show how our model can be transformed into an executable program with a usable interface, representing an important step towards deployment. We have undertaken a pilot evaluation in which a small number of lawyers have used the prototype program and given their feedback through a questionnaire. The feedback from the lawyers was very positive, showing that they are receptive to AI solutions that give effective, explainable decision support.