Recommendation Based Process Modeling Support: Method and User Experience

  • Authors:
  • Thomas Hornung;Agnes Koschmider;Georg Lausen

  • Affiliations:
  • Institute of Computer Science, Albert-Ludwigs University Freiburg, Germany;Institute of Applied Informatics and Formal Description Methods, Universität Karlsruhe (TH), Germany;Institute of Computer Science, Albert-Ludwigs University Freiburg, Germany

  • Venue:
  • ER '08 Proceedings of the 27th International Conference on Conceptual Modeling
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although most workflow management systems nowadays offer graphical editors for process modeling, the learning curve is still too steep for users who are unexperienced in process modeling. The efficiency of users may decrease when starting process modeling with minimal expertise and no obvious modeling support. This paper describes the first contribution towards a theoretically sound and empirically validated analysis of a recommender-based modeling support who is geared towards both novices and expert users. The idea is to interpret process descriptions as tags which describe the intention of the process. This leads us to the notion of virtual documents or signatures. Based on these signatures we provide a search interface to process models stored in a repository. Additionally the user can invoke a recommendation function during modeling time and the system automatically identifies and suggests relevant process fragments. By adding two additional criteria, the frequency of process reuse and structural correctness, we arrive at a full-fledged modeling support system, which provides an easy to use interface to the user while retaining a high fidelity to the user's modeling intentions. We validated our support system with a user experiment based on real-life process models and our prototype implementation.