Measuring similarity between semantic business process models

TLDR

Business process models can differ across modelers even when using the same language, and ontology-based descriptions are needed to resolve ambiguities from synonyms, homonyms, or varying abstraction levels, promising interoperability and interconnectivity. The study aims to automatically discover similar terms for process element names and facilitate semantic business process composition to support (semi-)automatic interoperability and interconnectivity. Solutions are based on an OWL DL‑based description of Petri nets, providing a formal framework for term discovery and composition.

Abstract

A business process may be modeled in different ways by different modelers even when utilizing the same modeling language. An appropriate method for solving ambiguity issues in process models caused by the use of synonyms, homonyms or different abstraction levels for process element names is the use of ontology-based descriptions of process models. So-called semantic business process models promise to support business process interoperability and interconnectivity. But, for (semi-) automatic process interoperability and interconnectivity two problems need to be solved. How can similar terms for process element names be automatically discovered and how can semantic business process composition be facilitated. In this paper we will present solutions for these problems based upon an OWL DL-based description of Petri nets.