N. Kama, T. French, and M. Reynolds (Australia)
Change impact analysis, change impact, requirement interactions, requirement, class interactions, class
Requirement interaction detection techniques aim to effectively detect the interactions between requirements in requirement artifacts. These techniques can be used to predict the class interactions by projecting the detected requirement interactions to class interactions that can be traced back to those requirements. However, the prediction tends to generate an excessive number of class interactions in some instances and not be able to detect some other (correct) class interactions. To improve the accuracy of the prediction, a set of filtration guidelines was used in the prediction process. This study evaluates the capability of the new filtration approach to predict the class interactions and the impact of the requirement interactions detection technique to the accuracy of the prediction results. The main contributions of the study are the demonstration that (1) the filtration approach improves the accuracy of the prediction by current approach and (2) the effectiveness of the requirement interactions detection technique to support the filtration approach in order to give high accuracy of prediction results.
Important Links:
Go Back
