Automated Generation of Consistent Graph Models with First-Order Logic Theorem Provers

Aren A Babikian,Oszkár Semeráth,Dániel Varró

doi:10.1007/978-3-030-45234-6_22

Aren A Babikian, Oszkár Semeráth + Show 1 more

Open Access

https://doi.org/10.1007/978-3-030-45234-6_22

Copy DOI

Abstract

The automated generation of graph models has become an enabler in several testing scenarios, including the testing of modeling environments used in the design of critical systems, or the synthesis of test contexts for autonomous vehicles. Those approaches rely on the automated construction of consistent graph models, where each model satisfies complex structural properties of the target domain captured in first-order logic predicates. In this paper, we propose a transformation technique to map such graph generation tasks to a problem consisting of first-order logic formulae, which can be solved by state-of-the-art TPTP-compliant theorem provers, producing valid graph models as outputs. We conducted performance measurements over all 73 theorem provers available in the TPTP library, and compared our approach with other solver-based approaches like Alloy and VIATRA Solver.

Highlights

Synthetic graph models have been in use for many challenges of software engineering including the testing of object-oriented programs [18, 20], quality assurance of domain-specific languages [28], validation of model transformations [7] or performance benchmarks of model repositories [5]
As an important technical side effect, thanks to a novel use of constants as object identifiers incorporated in the mapping to first-order logic (FOL) formulae, we managed to significantly improve the scalability of the Z3 SMT-solver for model generation purposes compared to existing approaches [28, 32], which relied upon the native support of decision procedures in SMT-solvers
We provided a mapping of domain-specific languages (DSL) specifications consisting of an EMF metamodel and well-formedness constraints into first-order logic formulae to be fed into Thousands of Problems for Theorem Provers (TPTP)-compliant theorem provers

Summary

Introduction

Synthetic graph models have been in use for many challenges of software engineering including the testing of object-oriented programs [18, 20], quality assurance of domain-specific languages [28], validation of model transformations [7] or performance benchmarks of model repositories [5]. Various lines of research in model-driven engineering rely upon such graph models. Network science heavily depends on the availability of graph models with designated distribution of nodes and edges. Active research in automated graph model generation [10,25,30,31] has been focusing on deriving graphs with desirable properties like consistency, diversity, scalability or realistic nature [37]. A challenging task of domainspecific model generators is to ensure consistency, i.e. to guarantee that synthetic models are compliant with the metamodel of the domain, but they satisfy additional well-formedness constraints captured in popular high-level languages like OCL or graph patterns

Objectives

Results

Conclusion