A Computational Ontological Model For Machine-Understandable Data in Artificial Intelligence

Big Data is believed to be one of the most important phenomena of the century due to its transformative effects and those undeniable effects of the data deluge in the fields of industry, science, and the Web -the fundamental wheels upon which today's world makes progress-are, however, hardly transformative. If Big Data is to lead to revolutions in these fields, the machine must be autonomous, that is, to perform higher-order cognitive skills, such as making decisions, inferences, and recommendations. The condition for an autonomous machine is its ability to understand and process Big Data. So, this work aims to determine the foundations for machine-understandability and accordingly determine the conditions for an autonomous machine. In this respect, this work proposes a machine ontology, Ontology 4.0, which represents phenomena in terms of their semantic properties in a relation-based fashion and processes semantic properties according to their types. Adapted from trope and type theories, this ontology forms the basis for machine-understandability. However, due to computational limitations from type theories, the semantic properties are discussed to be represented with nonextensional ontological objects, viz., urtropes, which turn out to be the building blocks of Ontology 4.0. Category theory is adapted as a formalization tool to implement this structure in the machine. Consequently, providing basis for machine-understandability, Ontology 4.0 is a machine ontology that harmonizes urtrope and modifed category theory.