Plugin de Alinhamento Interativo de Ontologias de Cibersegurança com Abordagens de Aprendizado de Máquina e K-Means

Technological advancements, combined with economic, political, and social transformations, have significantly contributed to the increasing complexity of Information Systems (IS). This complexity is associated with systems that integrate physical and cyber components through sensors, computational resources, and communication networks, commonly known as Cyber-Physical Systems (CPS). The rapid growth of CPS applications has brought challenges related to security, interoperability, and reliability, especially in critical infrastructures and sensitive functionalities.These challenges can be addressed through the use of ontologies in cybersecurity, which help organize and standardize knowledge to facilitate system integration. However, integrating these ontologies may expose vulnerabilities, allowing skilled attackers to exploit heterogeneity issues and inject malicious code. This study presents an Interactive Ontology Matching Plugin, based on the K-Means algorithm combined withthree machine learning classifiers, aiming to address heterogeneity problems in the integration of cybersecurity ontologies within nuclear power plant systems. This approach enables secure and efficient ontology matching, ensuring the continuous and protected operation of nuclear power plants against potential cyber-attacks.