Automatic Generation of Glossaries for Process Modelling Support

Abstract

Process models are often used for human to human communication. Besides other aspects, e.g., the chosen modelling notation or the model layout, the labelling has a strong influence on the understandability and, therefore, quality of a model. Consequently, labels should be reused and aligned across different process models. In order to support these goals, a glossary might be applied in the course of modelling. In this article, we argue that such a glossary can be generated automatically from the labels of an existing process model collection, e.g., a reference model. We introduce an approach for such a glossary generation that takes additional information on structural as well as control flow aspects into account. The applicability of our approach is illustrated by means of two case studies. Based thereon, we also report on findings regarding the appropriateness of the chosen structural and behavioural aspects.