Frankfurt, Hessen, Germany
32 days ago
Generative AI for Test Case Generation - Master Thesis Opportunity
Literature review: Conduct a comprehensive literature review on existing methods for automatic test case generation, machine learning, and generative AI. Identify state-of-the-art techniques and potential gaps in the current research.Data collection: Gather a representative dataset of Java and Perl codebases to serve as a basis for developing and evaluating the proposed strategies.Model selection: Experiment with various generative AI models, such as Transformers or Variational Autoencoders, and select the most suitable ones based on performance and feasibility. Develop prompting strategies.Evaluation: Evaluate the generated test cases' quality and coverage using metrics such as fault detection rate, mutation score, and branch coverage. Compare the results with those obtained from traditional manual testing methods.Strategy formulation: Based on the evaluation results, formulate strategies for efficiently generating test cases with high coverage for large codebases. Provide recommendations for integrating the proposed strategies into existing continuous integration/continuous deployment (CI/CD) pipelines.Presentation and dissemination: Present your findings in a final report and defend your thesis in front of a committee. Disseminate your results through academic publications, conference presentations, and workshops.
Confirm your E-mail: Send Email