Semantically Enabled Business Process Discovery

  1. Ibáñez, M.J. 1
  2. Vulcu, G. 2
  3. Ezpeleta, J. 1
  4. Bhiri, S. 2
  1. 1 Universidad de Zaragoza
    info

    Universidad de Zaragoza

    Zaragoza, España

    ROR https://ror.org/012a91z28

  2. 2 National University of Ireland
    info

    National University of Ireland

    Dublín, Irlanda

    ROR https://ror.org/00shsf120

Libro:
Proceedings of the ACM Symposium on Applied Computing

ISBN: 9781605586380

Año de publicación: 2010

Páginas: 1396-1403

Tipo: Capítulo de Libro

DOI: 10.1145/1774088.1774385 SCOPUS: 2-s2.0-77954697173 GOOGLE SCHOLAR

Resumen

Business process descriptions are usually stored in internal enterprise repositories. In order to be able to reuse Business Processes (a.k.a. BPs), BP designers require some tools to help them to discover processes (or fragments of processes) in the repository based on these descriptions. In most cases discovery is a difficult task (the diversity of modeling languages, the process descriptions are very close to the IT level being far from the business level, there is a lack of automatic tools, etc.). In this paper we investigate the use of semantics to alleviate the above mentioned problems providing with a method for the discovery of BPs. We have developed an RDF vocabulary to annotate and store BPs. First we use the vocabulary to annotate functional and non functional properties of basic activities of XML-based BP descriptions. Then we build an RDF knowledge base following the developed RDF vocabulary by extracting, in an automatic way, these properties and the structural properties from the BP description. In addition, functional and non functional properties of structured activities are automatically computed and added to the RDF knowledge base. Then the RDF knowledge base can be queried with SPARQL to achieve BPs discovery. In addition, we present an implementation prototype. © 2010 ACM.