A sketch of an approach for discovering UML use-case diagrams from textual specifications of systems using a chatbot

Abstract

Obtaining a model that describes the functionalities of a software program from a textual specification is a fundamental phase within the Model-Driven Architecture (MDA) proposed by the Object Management Group (OMG) that implements Model-Driven Engineering (MDE). A requirements model or Computation-Independent Model (CIM), which can be captured by a use case diagram in UML, is used to get the Platform-Independent Model (PIM) and then generate the source code of a software program. Various approaches have been suggested to discover a use-case diagram automatically or semi-automatically from a written specification. These methods employ predetermined rules to extract elements of the model, such as actors and use cases. However, these approaches place various constraints on how a specification is prepared. The advent of chatbots and their evident success in various natural language comprehension tasks, such as translation, text summarization, and sentiment analysis, hints at the potential for leveraging this conversational application to achieve our goal. In this paper, we explore the latest research in the field, and we sketch an approach using a chatbot to achieve our goal of extracting a use-case diagram from the textual specification of a software program. We focus on the experimentation of the tool for extracting a use-case diagram from a textual specification, representing one of the key components of our approach.