Foundations for multi-level ontology-based conceptual modeling

• Ano de defesa: 2016
• Autor(a) principal: Carvalho, Victorio Albani de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Doutorado em Ciência da Computação; Centro Tecnológico; UFES; Programa de Pós-Graduação em Informática
• Palavras-chave em Português: Power type; Conceptual modeling; Models; Ontology; Foundational ontology; Ontology-based conceptual modeling; Classification levels; Multi-level modeling; Modelagem conceitual; Ontologia de fundamentação; Modelagem conceitual baseada em ontologias; Níveis de classificação; Modelagem multi-nível; Ontologia; Modelos multinível (estatísticas); Ciência da Computação; 004
• Link de acesso: http://repositorio.ufes.br/handle/10/9870

Considering that conceptual models are produced with the aim of representing certain aspects of the physical and social world according to a specific conceptualization and that ontologies aim at describing conceptualizations, there has been growing interest in the use of ontologies to provide a sound theoretical basis for the discipline of conceptual modeling. This has given rise to a research area called ontology-based conceptual modeling, with significant advances to conceptual modeling in the last decades. Despite these advances, ontology-based conceptual modeling still lacks proper support to address subject domains that require not only the representation of categories of individuals but also the representation of categories of categories (or types of types). The representation of entities of multiple (related) classification “levels” has been the focus of a separate research area under the banner of multi-level modeling, aiming to address the limitations of the conventional two-level modeling paradigm. Despite the relevant contributions of multi-level modeling and ontology-based conceptual modeling, their combination has not yet received due attention. This work explores this gap by proposing the use of formal theories for multi-level modeling in combination with foundational ontologies to support what we call multi-level ontology-based conceptual modeling. To provide a well-founded approach to multi-level conceptual modeling, we develop a theory called MLT that formally characterizes the nature of classification levels and precisely defines the relations that may occur between elements of different classification levels. In order to leverage the benefits of the use of a foundational ontology to domains dealing with multiple classification levels, we combine the proposed multilevel modeling theory with a foundational ontology. This combination results in a hierarchical modeling approach that supports the construction of multi-level conceptual models in a spectrum of levels of specificity, from foundational ontologies to domain models. To demonstrate the applicability of our multi-level ontology-based conceptual modeling approach, we employ it to develop a core ontology for organizational structure, a domain that spans multiple classification levels. Further, we show how MLT can be used as a reference theory to clarify the semantics and enhance the expressiveness of UML with respect to the representation of multi-level models. The resulting UML profile enables the practical application of MLT.