Automatic Construction of OWL Ontologies From Petri Nets

Automatic Construction of OWL Ontologies From Petri Nets

Zongmin Ma (Nanjing University of Aeronautics and Astronautics, Nanjing, China), Haitao Cheng (Nanjing University of Posts and Telecommunications, Nanjing, China) and Li Yan (College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing, China)
Copyright: © 2019 |Pages: 31
DOI: 10.4018/IJSWIS.2019010102

Abstract

Ontology, as a formal representation method of domain knowledge, plays a particular important key role in semantic web. How to construct ontologies has become a key technology in the semantic web, especially constructing ontologies from existing domain knowledge. Currently, Petri nets have been a mathematical modeling tool, and have been widely studied and successfully applied in modeling of software engineering, database and artificial intelligence. In particular, PNML (Petri Net Markup Language) language has been a part of ISO/IEC Petri nets standard for representing and exchanging data on Petri nets. Therefore, how to construct ontologies from PNML model of Petri nets needs to be investigated. In this article, the authors investigate a method for automatic construction of web ontology language (OWL) ontologies from PNML of Petri nets. Firstly, this paper gives a formal definition and the semantics of PNML models of Petri nets. On this basis, a formal approach for constructing OWL ontologies from PNML model of Petri nets is proposed, i.e., this paper transforms Petri nets (including PNML model and PNML document of the Petri nets) into OWL ontologies at both structure and instance levels. Furthermore, the correctness of the transformation is proved. Finally, a prototype construction tool called PN2OWL is developed to transform Petri nets models into OWL ontologies automatically.
Article Preview

1. Introduction

The semantic web is an extension of the WWW, which goal is to create machine-processing information on the Web (Antoniou & van Harmelen, 2004). Ontology, which is a World Wide Web Consortium (W3C) standard knowledge representation model for the Semantic Web, allows the semantics of a domain to be expressed in a language understood by computers, enabling automatic processing of the meaning of shared information (Berners-Lee, Hendler, & Lassila, 2001). Full implementation of the Semantic Web requires widespread availability of ontologies and ontology construction has become a key problem to enable the Semantic Web (Maedche & Steffen, 2001). However, constructing ontologies is a difficult and time-consuming process. Automatic construction of ontologies is hereby an effective way to overcome ontology developing bottleneck. In recent years, a lot of studies has been dedicated to the construction of ontologies from diverse existing domains (Zhou, 2007).

As we have known, many domains are modeled by various data models, including relational data model, object-oriented data model, XML data model, ER/EER (entity-relationship/extended entity-relationship) data model, and UML (Unified Modeling Language) data model. Correspondingly, different approaches for constructing ontologies from various data models have been developed. Ontologies are automatically constructed, for example, by transforming relational databases (Lubyte and Tessaris, 2009), object-oriented databases (Zhang, Ma and Yan, 2011) and XML (Zhang and Ma, 2014). In addition, some conceptual data models are applied to automatically construct ontologies by transforming ER/EER data models (Han and Elmasri, 2003; Upadhyaya and Kumar, 2005; An, Mylopoulos and Borgida; 2006) and UML data models (Falkovych, Sabou and Stuckenschmidt, 2003; Xu et al., 2012).

In addition to various data models for domain modeling, Petri nets (Murata, 1989) have been proposed as a more precise way of modeling processes, which provide an abundance of analysis techniques to evaluate process modeling languages. Nowadays Petri nets have been widely investigated and successfully applied in many applications (e.g., mobile computing (Kahloul, Chaoui, & Djouani, 2015), smart homes (Nabih, Gomaa, Osman, & Aly, 2011), manufacturing systems (Chen, Li, Al-Ahmari, Wu, & Qu, 2017), business processes (Clempner, 2014; Kheldoun, Barkaoui, & Ioualalen, 2017), biology (Koch, 2015; Gogolinska, Jakubowski, & Nowak, 2016), and Web services (Cheng, Liu, Zhou, Zeng, & Ylä-Jääski, 2015)). Given the information resources of Petri nets in application domains, some efforts have been done to reuse/share Petri nets through the Semantic Web (Gašević & Devedžić, 2004). The approach in (Gašević & Devedžić, 2006) defined a Petri net ontology, which focuses on full semantic interoperability of Petri net models. In addition, Vidal, Lama, and Bugarin (2006) defined a high-level Petri net ontology. These two approaches above do not discuss the formal definition of Petri net and the correct proof of transformation rules, and do not implement a prototype construction tool. The mappings from Petri nets to OWL (Web Ontology Language) ontologies were established in (Zhang, Ma, and Ribaric, 2011), where the formal descriptions of Petri net models and OWL ontologies were presented but the comprehensive transformation rules and a prototype transformation tool were missing. A recent review of knowledge sharing between Petri nets and ontologies was presented in (Cheng & Ma, 2016).

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 16: 4 Issues (2020): Forthcoming, Available for Pre-Order
Volume 15: 4 Issues (2019): 3 Released, 1 Forthcoming
Volume 14: 4 Issues (2018)
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing