A comparison of NLP Tools for RE to extract Variation Points

In the requirement engineering of software product lines, several researches have focused on exploiting NLP techniques and tools to extract information related to features and variability from requirement documents. In a previous work we have proposed the use of the tool QuARS, a NLP Tool for Requirements Analysis, showing that some of the indicators used to detect NL ambiguity can also be exploited to detect variability. In this paper we discuss a comparison of the application at this regard of QuARS and other Requirements Analysis tools presented in the last edition of NLP4RE, in particular with respect to their ability to extract potential variation points, in search of better performances and of novel indicators.