Approximation methods in a software quality measurement framework

The basic approach in this paper is the presentation of a practical software quality measurement system that incorporates recent advances in rough set theory and parameterized approximation spaces in coping with the uncertainty in making decisions about software engineering data. This research also takes advantage of recent work on sensors, sensor fusion and signal analysis as well as software quality. The focus of this system is on the analysis of data from releases of a software product (e.g., gross changes, faults, design, size, complexity, and reuse data). A software quality measure quantifies the extent to which some specific attribute is present in a system, which is considered in the context of rough sets. An AR-scheme is a framework for information granule construction. An information granule is a clump of objects (points) drawn together by indistinguishability, similarity, or functionality. The basic idea is to set up a software quality measurement framework in the context of a parameterized approximation space to cope with the uncertainty, imprecision and incompleteness in software engineering data used to evaluate software quality.