Aggregating Measures using Fuzzy Logic for Evaluating Feature Models
Carla Ilane Moreira Bezerra, Rossana M. C. Andrade, José Maria Monteiro, Davi Cedraz · 2018
In the context of Software Product Lines (SPLs), evaluating the quality of a feature model is essential to ensure that errors in the early stages do not spread throughout the SPL. One way to evaluate a feature model is to use measures. However, measures alone are not enough to characterize the feature model quality, because most of them cover specific aspects, such as the number of features. So, there is a need for methods to aggregate measures at the level of quality sub-characteristic or characteristic. In this paper, we aim to investigate how to aggregate measures that have been proposed to evaluate the quality of feature models in SPL. We have used the fuzzy logic theory in order to aggregate these measures. The new aggregated measures can be applied to evaluate different and complex aspects of a feature model, such as: size, stability, flexibility and dynamicity. Moreover, to evaluate the use of the new aggregate measures, we applied them in different feature models. Our findings suggest that aggregate measures can assist the domain engineer in evaluating the maintainability of feature models.