Object-Oriented Data Flow Diagram Similarity Measurement Using Greedy Algorithm

Measuring software similarity base on software design artifacts has been one of the research topics in software engineering. Software similarity measurement is used in software reuse, clone detection, and learning assessment. Previous research focus on universal modeling language design artifacts. These design artifacts concern on the semantic and structure of a set of objects within a system and their interaction with themselves and objects outside the system. There are also design artifacts that model how data or information transformed into different forms, sizes, and time. This study focus on developing diagram similarity measurement base on one of those artifacts, i.e., object-oriented data flow diagrams. The measurement uses the greedy approach to calculate the similarity score of two diagrams using four elements derived from elements of object oriented data flow diagram, i.e. external entities, processes, data stores, and classes. The initial experimentation result shows that the proposed diagram similarity measurement is feasible.