Ambiguity in Requirements Engineering: Towards a Unifying Framework

Gervasi V., Ferrari A., Zowghi D., Spoletini P.

requirements engineering  ambiguity  natural language processing  Artificial Intelligence & Image Processing 

A long stream of research in RE has been devoted to analyzing the occurrences and consequences of ambiguity in requirements documents. Ambiguity often occurs in documents, most often in natural language (NL) ones, but occasionally also in formal specifications, be it because of abstraction, or of imprecise designation of which real-world entities are denotated by certain expressions. In many of those studies, ambiguity has been considered a defect to be avoided. In this paper, we investigate the nature of ambiguity, and advocate that the simplistic view of ambiguity as merely a defect in the document does not do justice to the complexity of this phenomenon. We offer a more extensive analysis, based on the multiple linguistic sources of ambiguity, and present a list of real-world cases, both in written matter and in oral interviews, that we analyze based on our framework. We hope that a better understanding of the phenomenon can help in the analysis of practical experiences and in the design of more effective methods to detect, mark and handle ambiguity.

Source: From Software Engineering to Formal Methods and Tools, and Back Essays Dedicated to Stefania Gnesi on the Occasion of Her 65th Birthday, edited by ter Beek M.H.; Fantechi A.; Semini L., pp. 191–210, 2019

Metrics