Evaluating automated support for requirements similarity analysis in market-driven development
Johan Natt och Dag, Björn Regnell, Pär Carlshamre, Michael Andersson, Joachim Karlsson · 2001
In market-driven software development, there is a considerable risk for congestion in the requirements engineering process, as the demand of short time-to-market is combined with a rapid arrival of new requirements from many different sources. Automated analysis of the continuous flow of incoming requirements provides an opportunity to increase the efficiency of the requirements engineering process. This paper presents empirical evaluations of the benefit of automated similarity analysis of textual requirements, where existing Information Retrieval techniques are used to statistically measure requirements similarity. The results show that automated analysis of similarity among textual requirements is a promising technique that may provide effective support in both requirements duplicate identification and requirements interdependency analysis.