SW-CASPAR: Reactive-cognitive architecture based on natural language processing for the task of decision-making in the open-world assumption

This paper addresses the issue of nowadays vocal assistants cognitive lacks, which are able to execute from vocal commands only simple plans without higher capabilities of decision-making. In this work we propose an open-world assumption transposition of the cognitive architecture CASPAR, whose heuristic takes into account of meta-reasoning in the closed-world assumption, namely SW-CASPAR. Such a cognitive architecture is also provided with a module for semi-automatic ontology learning from sentences in natural language, reflecting the domain with an instance of a novel foundational ontology called Linguistic Oriented Davidsonian Ontology (LODO), with the aim of increasing the deepness of reasoning without compromising linguistic-related features. LODO is inspired by the First-Order Logic Davidsonian notation and serialized in OWL 2. A case-study applied to automation on health scenarios is also provided.