Abstract

utilises semantic technologies (e.g., ontology) to model and query process representations. There are times in which such models must be reconstructed from existing textual documentation. In this scenario the automated generation of ontological models would be preferable, however current methods and technology are still not capable of automatically generating accurate semantic process models from textual descriptions. This research attempts to automate the process as much as possible by proposing a method that drives the transformation through the joint use of a foundational ontology and lexico-semantic analysis. The method is presented, demonstrated and evaluated. The original dataset represents 150 business activities related to the procurement processes of a case study company. As the evaluation shows, the proposed method can accurately map the linguistic patterns of the process descriptions to semantic patterns of the foundational ontology to a high level of accuracy; however further research is required in order to reduce the level of human intervention, expand the method so as to recognise further patterns of the foundational ontology and develop a tool to assist the business process modeller in the semi-automated generation of process models.