Justification, stability and relevance for case-based reasoning with incomplete focus cases

We define and study the notions of stability and relevance forprecedent-based reasoning, focusing on Horty’s result model ofprecedential constraint. According to this model, precedents con-strain the possible outcomes for a focus case, which is a yet unde-cided case, where precedents and the focus case are compared ontheir characteristics (called dimensions). In this paper, we refer tothe enforced outcome for the focus case as its justification status.In contrast to earlier work, we do not assume that all dimensionvalues of the focus case have been established with certainty: rather,each dimension is assigned a set of possible values. We define afocus case as stable if its justification status is the same for everychoice of the possible values. For focus cases that are not stable,we study the task of identifying relevance: which possible valuesshould be excluded to make the focus case stable? We show howthe tasks of identifying justification, stability and relevance canbe exploited for human-in-the-loop decision support. Finally, wediscuss the computational complexity of these tasks and provideefficient algorithms