Identifying of user interface elements from use case descriptions

Use cases are effectives ways to model the interaction between user and computer system, which covers not only how the user will interact with the system but how the system will respond to the user. They are generated from the requirements specification and are written in natural language narratives. This paper presents a prototype to identify user interface elements from use case descriptions using natural language processing techniques. We used the ANTLR tool, text patterns, part-of-speech tagger and stemming algorithms to resolve the problem. A preliminary assessement highlights promising results for the approach used.