XL-BPMN Model-Based Service Similarity Measurement Technique

Cheeyang Song, Eunsook Cho

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

In service-oriented developments, existing studies do not give lots of efforts on a formalized and systematic method for measuring similarities between services for their reuse in business models. This deteriorates the reusability of the constructed service due to the developers' intuition and informal service analyses. In this paper, we propose a technique for measuring similarity of services by analyzing syntax and semantics between services in the eXtended Layered business process modeling notation (XL-BPMN) model. First of all, the profiles of the formalized attributes for specifying services are defined, and the criteria for determining service similarities are established. To measure similarity between services, a technique both a syntactic similarity analysis facilitated by the XL-BPMN model-based edge counting method and a semantic similarity analysis based on meta data registry (MDR)-applied service attributes is specified. To automate analysis, a tool that can support the semantic similarity analysis technique is implemented. An online shopping mall system is investigated and evaluated to verify the effectiveness of the proposed technique. The similarity measurement technique, which is further formalized at upper business levels, can improve the accuracy of service analyses and enhance service reusability by distinguishing services with high similarity levels as common services.

Original languageEnglish
Pages (from-to)697-732
Number of pages36
JournalInternational Journal of Software Engineering and Knowledge Engineering
Volume33
Issue number5
DOIs
StatePublished - 1 May 2023

Keywords

  • BPMN
  • business process model
  • commonality
  • reuse service identification
  • Service similarity measures

Fingerprint

Dive into the research topics of 'XL-BPMN Model-Based Service Similarity Measurement Technique'. Together they form a unique fingerprint.

Cite this