Convergent synthesis of cognitive model based on deep learning and quantum semantics

Abstract

Cognitive modeling is the modeling enriched with semantic interpretations. Both denotative (formalized) and significative, or cognitive (non-formalized: subjective, mental, emotional, transcendental) semantics are taken into account. In digital data processing, the first ones are fairly well covered by discrete technologies, for example, big data analysis, while the second ones cannot be interpreted with discrete processes. Meanwhile, as shown in this paper, the subjective factor, although it is not explicitly formalized, can also be deterministically taken into account in cognitive modeling. This accounting can be implemented indirectly with specific information processing, for example, convergent decision making. It is also possible to emulate significative semantics through the use of heuristic algorithms, generated, for example, by a quantum mechanical or evolutionary approach. It is proposed to carry out an automated synthesis of cognitive models with an in-depth cognitive interpretation of model components. This is achieved by applying the author's convergent approach to decision support. It is experimentally shown the high quality level of automatic recognition and synthesis of elements of cognitive models (accuracy - 93%).