Towards statistical comparison and analysis of models

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

7 Citations (Scopus)
99 Downloads (Pure)

Abstract

Model comparison is an important challenge in model-driven engineering, with many application areas such as model versioning and domain model recovery. There are numerous techniques that address this challenge in the literature, ranging from graph-based to linguistic ones. Most of these involve pairwise comparison, which might work, e.g. for model versioning with a small number of models to consider. However, they mostly ignore the case where there is a large number of models to compare, such as in common domain model/metamodel recovery from multiple models. In this paper we present a generic approach for model comparison and analysis as an exploratory first step for model recovery. We propose representing models in vector space model, and applying clustering techniques to compare and analyse a large set of models. We demonstrate our approach on a synthetic dataset of models generated via genetic algorithms.
Original languageEnglish
Title of host publicationProceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy
Pages361-367
DOIs
Publication statusPublished - 2016
Event4th International Conference on Model-Driven Engineering and Software Development (MODELSWARD 2016) - Rome, Italy
Duration: 19 Feb 201621 Feb 2016
http://www.modelsward.org/?y=2016

Conference

Conference4th International Conference on Model-Driven Engineering and Software Development (MODELSWARD 2016)
Abbreviated titleMODELSWARD 2016
CountryItaly
CityRome
Period19/02/1621/02/16
Internet address

Fingerprint

Recovery
Vector spaces
Linguistics
Genetic algorithms

Cite this

Babur, Ö., Cleophas, L., Verhoeff, T., & van den Brand, M. (2016). Towards statistical comparison and analysis of models. In Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy (pp. 361-367) https://doi.org/10.5220/0005799103610367
Babur, Ö. ; Cleophas, L. ; Verhoeff, T. ; van den Brand, M. / Towards statistical comparison and analysis of models. Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy . 2016. pp. 361-367
@inbook{004b490cab4349788a232143a807bdf5,
title = "Towards statistical comparison and analysis of models",
abstract = "Model comparison is an important challenge in model-driven engineering, with many application areas such as model versioning and domain model recovery. There are numerous techniques that address this challenge in the literature, ranging from graph-based to linguistic ones. Most of these involve pairwise comparison, which might work, e.g. for model versioning with a small number of models to consider. However, they mostly ignore the case where there is a large number of models to compare, such as in common domain model/metamodel recovery from multiple models. In this paper we present a generic approach for model comparison and analysis as an exploratory first step for model recovery. We propose representing models in vector space model, and applying clustering techniques to compare and analyse a large set of models. We demonstrate our approach on a synthetic dataset of models generated via genetic algorithms.",
author = "{\"O}. Babur and L. Cleophas and T. Verhoeff and {van den Brand}, M.",
year = "2016",
doi = "10.5220/0005799103610367",
language = "English",
isbn = "978-989-728-168-7",
pages = "361--367",
booktitle = "Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy",

}

Babur, Ö, Cleophas, L, Verhoeff, T & van den Brand, M 2016, Towards statistical comparison and analysis of models. in Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy . pp. 361-367, 4th International Conference on Model-Driven Engineering and Software Development (MODELSWARD 2016), Rome, Italy, 19/02/16. https://doi.org/10.5220/0005799103610367

Towards statistical comparison and analysis of models. / Babur, Ö.; Cleophas, L.; Verhoeff, T.; van den Brand, M.

Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy . 2016. p. 361-367.

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

TY - CHAP

T1 - Towards statistical comparison and analysis of models

AU - Babur, Ö.

AU - Cleophas, L.

AU - Verhoeff, T.

AU - van den Brand, M.

PY - 2016

Y1 - 2016

N2 - Model comparison is an important challenge in model-driven engineering, with many application areas such as model versioning and domain model recovery. There are numerous techniques that address this challenge in the literature, ranging from graph-based to linguistic ones. Most of these involve pairwise comparison, which might work, e.g. for model versioning with a small number of models to consider. However, they mostly ignore the case where there is a large number of models to compare, such as in common domain model/metamodel recovery from multiple models. In this paper we present a generic approach for model comparison and analysis as an exploratory first step for model recovery. We propose representing models in vector space model, and applying clustering techniques to compare and analyse a large set of models. We demonstrate our approach on a synthetic dataset of models generated via genetic algorithms.

AB - Model comparison is an important challenge in model-driven engineering, with many application areas such as model versioning and domain model recovery. There are numerous techniques that address this challenge in the literature, ranging from graph-based to linguistic ones. Most of these involve pairwise comparison, which might work, e.g. for model versioning with a small number of models to consider. However, they mostly ignore the case where there is a large number of models to compare, such as in common domain model/metamodel recovery from multiple models. In this paper we present a generic approach for model comparison and analysis as an exploratory first step for model recovery. We propose representing models in vector space model, and applying clustering techniques to compare and analyse a large set of models. We demonstrate our approach on a synthetic dataset of models generated via genetic algorithms.

U2 - 10.5220/0005799103610367

DO - 10.5220/0005799103610367

M3 - Chapter

SN - 978-989-728-168-7

SP - 361

EP - 367

BT - Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy

ER -

Babur Ö, Cleophas L, Verhoeff T, van den Brand M. Towards statistical comparison and analysis of models. In Proceedings of the 4th International Conference on Model-Driven Engineering and Software Development, February 19-21, 2016, in Rome, Italy . 2016. p. 361-367 https://doi.org/10.5220/0005799103610367