Controlled Experiment in Crosscutting Concerns Identification from Software Requirements Specification

<p>The concern is a group of requirements with the same purpose, which are known as crosscutting concerns when they are scattered and tangled in the system. Identifying and separating these concerns is a matter of huge importance to software maintainability and evolution. For this, some approaches are proposed. There are few experimental studies comparing and analyzing these approaches in detail. In this paper, our aim is to provide empirical evidence about two approaches in this context. We conducted a controlled experiment to compare the effectiveness of two approaches, which identify crosscutting concerns in the requirements specification, followed by an assessment of its perceived utility and ease of use. Our results indicate that, in this given context and planned scenario, there are significant differences between the effectiveness of ObasCId and Theme/Doc approaches.</p>