Razvoj inteligentnih sistema vođen domenskim ontologijama
Tošić, Milorad B., 1969
Stojanović, Dragan H., 1977-
Nejković, Valentina, 1975-
Bogdanović, Miloš, 1959-
Devedžić, Vladan
Systems based on digital devices and intelligent computer technology are of utmost importance across various domains starting from entertainment, through business and banking to industry, manufacturing and healthcare. However, due to their increasing complexity, heterogeneity and rising number, the operations related to management, control and coordination within these systems become significantly more demanding, requiring much time and effort when done entirely manually. Apart from that, it is often the case that detailed domain expertise about the devices as well as underlying processes is necessary, which increases the overall cognitive load while handling such systems. Automated code generation is one of the most promising directions to overcome the mentioned challgenges.In this dissertation, an ontology-based framework is introduced for automated code generation, with aim to make the development and management of complex intelligent systems more convenient for adoption within heterogeneous domains. Ontologies are leveraged for representation of knowledge covering crucial aspects of intelligent systems, including domain conceptualization, topology and structure together with system behavior. Starting from semantic knowledge representation with respect to ontologies in synergy with automated code generation, auxiliary tools which enable the adoption of intuitive domain-specific notation are created as outcome with goal of automated management and development of intelligent systems.Effectivenss of the proposed approach is illustrated based on experiments within realistic case studies from various domains of usage: fog computing, experimental robotics, blockchain systems and augmented reality. When it comes to evaluation, many relevant points of view are considered, such as execution time, performance improvement or speed-up compared to more traditional approaches. According to the achieved results, the proposed approach significantly speeds up handling, development and maintenance of intelligent systems and applications. Apart from that, based on semantic knowledge representation, this approach exhibits many additional benefits, such as context-aware adaptive system behavior, coordination and verification of relevant elements, as illustrated by presented case studies.
Biografija autora: list 172.Bibliografija: list. 151-163. Datum odbrane: 04.02.2025 Knowledge representation
srpski
2024
Ovo delo je licencirano pod uslovima licence
Creative Commons CC BY-NC-ND 3.0 AT - Creative Commons Autorstvo - Nekomercijalno - Bez prerada 3.0 Austria License.
http://creativecommons.org/licenses/by-nc-nd/3.0/at/legalcode