Reflection of expert-based evaluation of software quality

Abstract

The method of reflections data, which is obtained as a result of processing expert-based evaluation of software quality by different criteria of evaluation using petal charts. Reflection of expert-based evaluation of software quality is found to be the process of reflections data in graphical form for maximum ease of understanding and rapid perception, and also providing a clear overview and shape of any object, process or phenomenon. The software quality evaluation criteria and their weighting factors for each expert are offered, which provide a reliable presentation of the current state of the software development process, a proper understanding of the problems that may arise at any stage of the program project implementation, and the exact characteristics of their components. This mechanism of data reflection enables business analysts to submit qualitatively and quantitatively multiple values of complex software quality indicators that can be obtained from the results of any survey of different experts at a particular stage of the program project implementation. The proposed methodology is suitable for presenting a variety of expert survey results, with a subdivision into unlimited number of the roles of the software quality evaluation participants concerning the importance of each of them. An algorithm for calculating the area of sectorial petals in a polar coordinate system has been developed, using which it is possible to calculate and evaluate the relative software quality according to the proper criteria. The final complex software quality indicators for each expert are determined, and a comprehensive indicator of its quality is summarized as well. An algorithm for calculating the area of a petal diagram in a polar coordinate system has been developed, which allows determining the part of software quality according to all the criteria, which is currently evaluated by one of the experts, as well as the part of software quality, which has yet to be achieved for its 100% completeness. The relevant conclusions are drawn and recommendations are made concerning the use of the developed data reflection methodology.