Symbolic Approximate Reasoning Within Unbalanced Multi-sets: Application to Autism Diagnosis

In most daily activities, humans often use imprecise information derived from appreciation instead of exact measurements to make decisions. Multisets allow the representation of imperfect information in a Knowledge-Based System (KBS), in the multivalued logic context. New facts are deduced using approximate reasoning. In the literature, dealing with imperfect information relies on an implicit assumption: the distribution of terms is uniform on a scale ranging from 0 to 1. Nevertheless, in some cases, a sub-domain of this scale may be more informative and may include more terms. In this work, we focus on approximate reasoning within these sets, known as unbalanced sets, in the context of multi-valued logic. We introduce an approach based on the Generalized Modus Ponens (GMP) model using Generalized Symbolic Modifiers (GSM). The proposed model is implemented in a tool for autism diagnosis by means of unbalanced severity degrees of the Childhood Autism Rating Scale (CARS). We obtain satisfying results on the distinction between autistic and not autistic child compared to psychiatrists diagnosis.