An Empirical Study on Assessing the Quality of Use Case Metrics

Use cases are generally meant to describe the functional requirements of a software system. However, the use of some Natural Language (NL) text may inherently introduce language and interpretation related issues. Several tools and techniques have been proposed and available to assess the quality of use case specification, however, often performed manually. The precise and automated way of analyzing the quality of use cases in different aspects is a need due to volatile functionalities and rapid change in requirements. In this paper, we report the results of two separate experimental studies conducted, a replication of one another, to evaluate the significance and relevance of the use case quality assessment metrics. Our results revealed redundancies among the parameters associated with the quality measures and suggested modifications on the formulation of use case metrics which in turn make them complete, correct and consistent. Subsequently, we develop a tool support to automatically analyze the quality of use case specification on the basis of experimentally validated metrics.