Generating UML Use Case and Activity Diagrams Using NLP Techniques and Heuristics Rules

Abstract

The process of generating Unified Modeling Language (UML) Diagrams from Natural Language (NL) requirements is considered a complex and challenging task. Software requirements specification is often written in NL format, which causes potential problems. Requirement's analysts analyze and process natural language requirements manually to extract the UML elements. The manual analysis takes a lot of time and effort, which justifies the need for automated support. This paper proposes an approach to facilitate the NL requirements analysis process and UML diagrams extraction from NL textual requirements using natural language processing (NLP) techniques and heuristics rules. This approach focuses on generating use-case and activity diagrams. The approach has been applied to a case study and evaluated through an experimental. The evaluation of the approach will be conducted through a comparative study. The experimental results prove that the proposed approach is considerably improved as compared to the other approaches.