O. Garcia, S.V. Nagarajan, and P. Croll (Australia)
Automatic testing, quality model, object-orientedprograms and object-oriented metrics.
Defining and measuring software quality has long been a research topic in Software Engineering; in fact quality was the driving force behind the birth of Software Engineering. The golden goal of any software quality model is to get a sensible measure of the overall quality of software products and to be able to compare quality among similar products. Most of the quality models available now cannot be computed automatically, even less without intense human participation. Having been lecturers for many years it would be very useful to have an automatic marking system that will be more reliable, precise and assertive than the manual process we have used so far. Even though we are not there yet, this paper will show our journey on that path and show some of the results we have achieved so far. Our contributions in this paper are (1) To present an integral quality model for small object-oriented programs (2) To show some preliminary results of applying this model to a sample of second year Java assignments (3) To elaborate how this model can be computed automatically (4) To present the limitations and challenges that this approach faces.
Important Links:
Go Back
