Test Case Generation using Graph-based Deep Learning Methods

In today's fast-paced software development environment, the need for automated and reliable test case generation is increasingly critical. Traditional methods often fail to ensure both syntactic accuracy and semantic relevance, especially in complex domains. This paper proposes a hybrid approach that combines graph-based deep learning with large language models (LLMs) to generate test cases automatically. By incorporating both semantic and syntactic analysis, the system will produce context aware, structurally sound test cases, adaptable across diverse software domains. This solution aims to improve testing efficiency and coverage, ensuring higher-quality software development and streamlined testing processes.