Automation and Visualization of Knowledge in Cloud Services: A Semantic Ontology Enhanced by Generative AI Models

The rapid evolution of Cloud Computing has revolutionized the delivery and use of IT services. However, challenges such as portability and interoperability among cloud platforms hinder seamless service integration. This thesis addresses these issues by proposing a conceptual framework for the semantic representation and enrichment of a Cloud Ontology, supported by Artificial Intelligence (AI) technologies. The work focuses on leveraging Large Language Models (LLM) and Retrieval-Augmented Generation (RAG) techniques to automate the population and management of ontologies. Advanced tools such as Protégé, Dify, and RAGFlow were utilized to develop innovative solutions for representing, querying, and interactively visualizing knowledge in cloud domains. Semantic technologies like RDF and OWL ensured broad interoperability, while the integration of linguistic models enabled continuous and dynamic updates of information. The results demonstrate significant improvements in the automation of ontology management processes and access to structured data. This thesis contributes to the field of Knowledge Engineering by proposing an integrated approach to tackling the complexity and fragmentation of the cloud landscape. Finally, future perspectives are outlined for applying these techniques in broader contexts, such as big data management and cross-industry interoperability.