Modeling interactions using role-driven patterns

Great interest in model-driven approaches that are based on the automatic transformation of models has emerged. In this software development paradigm, the models and their transformations are specified at a high abstraction level supporting evolution, refinement and code generation. One of the benefits that these approaches try to achieve is improving the software quality through the reusability of well proven patterns. In order to achieve this goal, this paper presents an approach that defines patterns to transform use case models into interaction models. This approach uses roles to capture the linguistic abstractions on the use case text as well as to describe the corresponding interaction. The roles are defined as model elements to guarantee genericity. In this way, the transformation patterns are independent of a specific domain or a technological platform. The defined patterns have been validated. Some of the validation results are also presented in this paper.