HermieLab is a framework to automatically generate LearnLib Automata Learning Experiment Setups. With these, we can automatically reverse engineer complex programs to automaton-based representations of their behaviour.
It all started as a bachelor thesis written by Hans van der Laan under the supervision of prof. Arend Rensink. The goal of this project was to use automata learning as a technique to learn behavioral models of GUI-based programs. Both intrigued by the results, we decided to work further on it. Almost a year of hard work later, the HermieLab framework was born!
For more information about this framework, please check out our wiki!