Automated Assessment of Use Cases Elicitation from Requirements Text

Software Requirements Engineering addresses specific challenges which exist in the effort to gain an understanding of the nature of the engineering problem arising from user’s real-world needs and desires. This research is aimed at helping software analysts meet these challenges. The proposed methodology forms the basis of the automated process designed to capture the high-level system services and actors from the textual user requirements. This model is intended to serve as a basis for software Use-Case Model development, and can be used by analysts in their in-depth study of requirements text. The approach is rooted in the syntactical analysis and formalization of text written in natural language, and it is enriched with domain-related information provided by the Expert Comparable Contextual (ECC) models that are extracted from reusable domain-specific data models. We illustrate the applicability of our methodology on an order invoicing case study and demonstrate it with a prototype tool. The results of the validation of our methodology prove that such a tool for assisting the elicitation of use-case models from textual requirements is feasible.