Publication

Abstract : As we know in today’s modern software engineering, use case diagrams for representing the functionalities of the system act significantly in mediating the gap between the stakeholder and the developer. Some of the earlier techniques used in constructing such diagrams can be problematic when these diagrams are to be employed in large systems where there are many actors and activities. That is why every limitation outlined above might be resolved further with the help of NLP and developments in graphs theory. The proposed method makes improvements by reducing the human interference; it can handle increase in system size and the number of actors and their interactions; it also provides highly accurate shortest-path identification through use of validation methods. In addition, shortest-path algorithms are incorporated to improve the validation exercise, thereby minimizing errors common in presenting the system functionalities. Hence, the results indicate that the proposed method is more effective not only for generating use case diagrams with increased precision and scalability but also for constructing the necessary framework for checking system interactions’ correctness. This work provides a foundation for future studies aimed at enhancing the automation of the use case diagram especially for complex systems.