On Building an Integrated and Generic Platform for Software Quality Evaluation

摘要

In this paper, we introduce an approach and tools dealing with the software quality modelling and evaluation. The approach is based on the use of both the model transformation and the machine learning. We first define a software quality ontology formalised by an UML class diagram. The ontology serves then as a quality meta-model that must be instantiated to produce quality models for specific domains or projects. The instantiation is achieved by the combination of the model transformation implemented by graph rewriting rules and the machine learning aiming at discover specific relationships relating the software quality concepts. The approach is implemented by a platform based on the use of GXL(graph exchange language) and AGG that is a graph rewriting system