Revisiting AI and Testing Methods to Infer FSM Models of Black-Box Systems
- Resource Type
- Conference
- Authors
- Groz, Roland; Simao, Adenilso; Bremond, Nicolas; Oriat, Catherine
- Source
- 2018 IEEE/ACM 13th International Workshop on Automation of Software Test (AST) AST Automation of Software Test (AST), 2018 IEEE/ACM 13th International Workshop on. :16-19 May, 2018
- Subject
- Computing and Processing
Testing
Inference algorithms
Software
Software algorithms
Automata
Approximation algorithms
Artificial intelligence
Query learning
FSM Testing
Software Engineering
- Language
Machine learning in the form of inference of state machine models has gained popularity in model-based testing as a means of retrieving models from software systems. By combining an old idea from machine inference with methods from automata testing in a heuristic approach, we propose a new promising direction for inferring black box systems that cannot be reset. Preliminary experiments show that this heuristic approach scales up well and outperforms more systematic approaches.