Resumo

Intelligent systems are algorithm-built models that enhance interdisciplinary strategies and broader utility. However, system development faces challenges in effective decision-making, data management, integration, quality, security, and resource optimization. Information systems often fail to facilitate decision-making due to fragmentation, isolated initiatives, and lack of effectiveness and integration. Frameworks offer a solution by providing practices, methods, and tools for problem-solving and improving information sharing and decision-making.

How can neuro-symbolic A.I. and neural network components be integrated into administrative systems employing a framework? It is worth discussing the relevance of a structure incorporating neuro-symbolic A.I. and neural network components to improve information systems' usability, efficiency, and effectiveness. A deeper understanding of the interaction between humans and computers, as a function of social and cognitive aspects, should be appreciated, making systems suitable and relevant for users.

Intelligent agents are decentralized pattern-grasping (machine learning) mechanisms that allow critical situations to be anticipated and the causes, reasons, and possible outcomes for operational problems to be accurately identified, even when no experts are available in the organization. Optimizing the administrative routine and the response time in the management support processes allows the worker more quality and agility in answering user requests, with immediate and secure access to databases.

The keys to succeed in transformative technology include improving information competencies at all levels of the organization. Frameworks must go beyond programming logic to incorporate new knowledge in cross-layer learning, investigating how new knowledge can be represented as meaningful signals and incorporated into the process. The framework provides a structured approach to integrating neuro-symbolic A.I. and neural network components in administrative systems, assimilating the methodologies for identifying, analyzing, and proposing solutions to specific problems.

By effectively combining people, processes, and organizations while considering the inherent subjectivity of decision-making, the proposed framework provides a comprehensive approach to improve efficiency, accuracy, compatibility, scalability, interoperability, and decision-making capabilities. The framework proves particularly valuable in tasks involving sequential or time-dependent data. It aids in planning supervised learning, evaluating responsiveness during training, employing unsupervised learning techniques to uncover patterns, and identifying ways to understand user behaviors.

BENBYA, H.; DAVENPORT, T. H.; PACHIDI, S. Artificial intelligence in organizations: Current state and future opportunities. MIS Quarterly Executive, v. 19, n. 4, 2020. BIDOLLAHKHANI, M.; ATASOY, F.; ABDELLATEF, H. LTC-SE: Expanding the Potential of Liquid Time-Constant Neural Networks for Scalable AI and Embedded Systems. arXiv preprint. 2023. BOUNEFFOUF, D.; AGGARWAL, C. C. Survey on Applications of Neurosymbolic Artificial Intelligence. arXiv preprint. 2022 DEMIDOVSKIJ, A. V.; BABKIN, E. A. Integrated neurosymbolic decision support systems: problems and opportunities. Business Informatics. v