Avaliação da modelagem conceitual de sistemas de informação a partir de ontologias de fundamentação: verificação de relações parte-todo

Abstract

Software development companies have worked in a highly competitive and dynamic market. One can easily realize the need of increasing investments in activities of software engineering, as an attempt to improve the processes of development and ultimately to provide better quality products and services. In order to cope with the own needs for which the information systems were planned, as well as to improve the overall quality of the product for the final user, those information systems should be based on consistent representations of the knowledge domain in which they will operate. Within this context, the conceptual modeling activities have an important representational role. Models built according an ad-hoc way, without guidelines to regulate the actions of modeling, can generate different sorts of problems: models with poor representation of the reality, difficulties in communication, and impossibility of interoperability among systems. We present here a research that consists in the evaluation of conceptual models using a foundational ontology. We checked real conceptual models represented in Unified Modeling Language (UML), which are part of a project of large information systems of the government of the state of Minas Gerais (Brazil), regarding their adherence to ontological criteria. In order to perform this evaluation, we employed a foundational ontology named Unified Foundational Ontology (UFO) and, in particular, a modeling language proposed by that ontology, namely, OntoUML Lightweight Editor (OLED). We propose new improved models using these procedures based on ontological criteria based on the changes in part-whole relations. Eventually, we found proper semantic models from an ontological point of view. We hope our research can contribute to the improvement of conceptual modeling activities and for the development of information systems able to properly represent the reality. In addition, we hope that this initiative be brought more closely information science and computer science through the theories of ontology.