Automatic Test Case Generation Using UML Diagrams and User Stories: An AI-Powered Approach

Testing software is one of the most important steps in building reliable applications. However, writing test cases by hand takes a lot of time and is often incomplete. This paper presents an AI-powered system that automatically generates test cases by combining two inputs: UML diagrams (blueprints of how the software is built) and User Stories (plain English descriptions of what the software should do). The system uses spaCy for NLP-based User Story parsing, Neo4j as a Knowledge Graph database, LangChain to orchestrate prompts, and Groq Llama 3 70B to generate tests. The Groq Llama 4 Scout vision model allows users to upload hand-drawn UML diagrams as images. Results show that combining both inputs achieves High test coverage and High edge-case detection — significantly better than tools that rely on only one source, which achieve only Medium or Low levels.

Index Terms— Automated Test Case Ge

M.Nagamma, Siva Sankar M, Magesh R, Yuvan U (2026). Automatic Test Case Generation Using UML Diagrams and User Stories: An AI-Powered Approach. International Journal of Innovative Research in Multidisciplinary Education & Technology (IJIRMET), 11(3).