A goal-question-metrics model for configuration knowledge bases

Configuration knowledge bases are a wellestablished technology for describing configurable products like cars, computers, and financial services. Such knowledge bases are characterized by sets of constraints, variables, and domains. Lot of research has been done for testing knowledge bases, finding conflicts, and recommending repair actions. In contrast, less work has been done in the area of measuring the quality of configuration knowledge bases. Such quality measurements can help knowledge engineers to improve the maintainability, understandability, and functionality of knowledge bases. Based on a literature review we first give an overview of the state-of-the-art in knowledge base metrics. We will extend the current research by using the goal-questionmetrics (GQM) approach of the software engineering discipline to find gaps for the characterization of knowledge bases. We will also identify further metrics to complete the model. The results of this paper help knowledge engineers to reduce the effort to develop and maintain configuration knowledge bases.