Measuring structural complexity for class diagrams: an information theory approach

  • Authors:
  • Yuming Zhou;Baowen Xu

  • Affiliations:
  • Southeast University, Nanjing, P.R. China;Southeast University, Nanjing, P.R. China

  • Venue:
  • Proceedings of the 2005 ACM symposium on Applied computing
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the object-oriented analysis phase, class diagrams are used to represent the static structure of objects in a system and the kinds of static relationships that exist among them, which lay the foundation for all later design and development. Therefore, their complexity has a significant impact on the quality of the ultimate system. For quantifying the structural complexity of a class diagram precisely, this paper presents a structural complexity model called weighted class dependence graph (WCDG), in which almost all relationships among classes/interfaces that can be found during the object-oriented analysis phase are represented and their degrees are characterized by weights. Based on the WCDG, this paper depicts the occurrence of the outgoing and incoming edges by two discrete random variables and then uses entropy distance to derive a novel structural complexity measure called EDC. We believe that EDC is flexible and could evaluate the structural complexity of class diagrams more objectively.