A Term-based Approach for Generating Finite Automata from Interaction Diagrams

Non-deterministic Finite Automata (NFA) may represent regular languages succinctly (in number of states), increasing their appeal for applications such as word recognition. This paper proposes a new approach to generate NFA from an interaction language such as UML Sequence Diagrams or Message Sequence Charts. Using an operational semantics, we generate an NFA from a set of interactions reachable using the associated execution relation. In addition, by applying simplifications (rewriting) on reachable interactions to merge them, it is possible to obtain reduced NFA without relying on costly NFA reduction techniques. Experimental results regarding NFA generation and their application in trace analysis are also presented.