Building Quality Estimation models with Fuzzy Threshold Values

This work presents an approach to circumvent one of the major problems with techniques to build and apply software quality estimation models, namely the use of precise metric thresholds values. We used a fuzzy logic based approach to investigate the stability of a reusable class library interface, using structural metrics as stability indicators. To evaluate this new approach, we conducted a study on three versions of a commercial C++ class library. The obtained results are very promising when compared to those of two classical machine learning (ML) approaches, Top Down Induction of Decision Trees and Bayesian classifiers.