Resumen de Midas: a model-driven approach for context-aware applications
Users of computer systems are becoming more demanding and require solutions highly tailored to their special requirements. Therefore, context-aware applications are a promising approach because they use context to provide the users with relevant information or services.
Generally, requirements and behavior of context-aware applications are directly specified in the source code, and their implementation is performed in a non-automatic fashion. As a consequence, these kinds of applications tend to be complex to develop.
The hypothesis of this dissertation is that a model-driven approach can be used to specify, in an abstract way, the requirements and behavior of context-aware applications, and to translate requirements specifications and behavioral models to implementation models. Thus, software developers should not worry about issues inherent to the solution domain, and the development process of context-aware applications will be easier tan the traditional.
The core contribution of this dissertation is Midas, a model-driven approach to automate the development of context-aware applications. Midas comprises a domain-specific language to model, at a higher level of abstraction, requirements and behavior; and a set of transformations to derive implementation models from the requirement specifications and the behavioral models.
The approach was applied on two case studies. The results of the research performed indicate that the proposed language addresses the representation of requirements and behavior of context-aware applications, and that the transformations translate requirements specifications and behavioral models to implementation models.
