Model based testing with machine learning techniques

This project aims to assess the application of active and passive machine learning techniques for testing complex software systems.
Master

We are interested in a variety of topics in the applications of intelligent methods (e.g., machine learning and search algorithms) for engineering complex software systems. Example of these topics include:

  • Using model execution and machine learning techniques (e.g., reinforcement learning) to cost-effectively test Cyber-Physical Systems (CPSs) or IoT in the presence of environmental uncertainty.
  • Testing machine learning algorithms in software systems such as embedded in autonomous vehicles.
  • Testing security and privacy of critical infrastructures, e.g., used in oil & gas, healthcare welfare technologies, and cancer registry system.

Goal:

Developing software testing tools for the above topics. There is a possibility to apply the tools to the real case studies.

Learning outcome:

  • Software Testing
  • Machine Learning (e.g., active and passive learning)
  • Search Algorithms (e.g., Genetic Algorithms)
  • Model-based Testing

Supervisors

  • Shaukat Ali
  • Tao Yue