Article
Journal :
Malaysian Journal of Computer Science
ISSN : 0127-9084
Publisher :
University of Malaya
Information
Period : January 2021
Volume : 34 Number : 1
Pages : 13–33
Details
An Approach And A Tool For Merging A Set Of Models In Pairwise Way
Model merging addresses the problem of combining information from a set of models into a single one. This task is considered to be an important step in various software engineering practices. When many (more than two) models need to be merged, the most practiced technique is to perform the merge in a pairwise way, without considering the order of merging. The problem with this technique is that the resulting quality is not guaranteed because it is influenced by such an order. In this paper, we propose a pairwise approach for model merging aiming to provide better results by taking into account the order of merging. This approach proposes to combine models in an iterative process until obtaining only one model. At each iteration, we first compare each pair of models in order to measure the similarity between them and to identify the correspondences between their component elements. This is performed using two heuristic-based operators respectively named compare and match. After that, we identify the most similar pairs of models and merge them using a proposed operator. We have implemented our approach as tool support called 3M and evaluated it on a set of case studies.
Key words :
Model comparison Maximum weighted matching Compare Match Merge Model merging
Ref. laboratory citation :
misc-lab-351
Link :
Texte intégral
ACM :
M. Boubakir and A. Chaoui. 2021. An Approach And A Tool For Merging A Set Of Models In Pairwise Way. Malaysian Journal of Computer Science, 34, 1 (January 2021), University of Malaya, 13–33. https://ejournal.um.edu.my/index.php/MJCS/article/view/12728.
APA :
Boubakir, M. & Chaoui, A. (2021, January). An Approach And A Tool For Merging A Set Of Models In Pairwise Way. Malaysian Journal of Computer Science, 34(1), University of Malaya, 13–33.Retrieved from https://ejournal.um.edu.my/index.php/MJCS/article/view/12728
IEEE :
M. Boubakir and A. Chaoui, "An Approach And A Tool For Merging A Set Of Models In Pairwise Way". Malaysian Journal of Computer Science, vol. 34, no. 1, University of Malaya, pp. 13–33, January, 2021, https://ejournal.um.edu.my/index.php/MJCS/article/view/12728.
BibTeX :
@article{misc-lab-351,
author = {Boubakir, Mohamed and Chaoui, Allaoua},
title = {An Approach And A Tool For Merging A Set Of Models In Pairwise Way},
journal = {Malaysian Journal of Computer Science},
volume = {34},
number = {1},
issn = {0127-9084},
pages = {13–33},
publisher = {University of Malaya},
year = {2021},
month = {January},
url = {https://ejournal.um.edu.my/index.php/MJCS/article/view/12728},
keywords = {Model comparison, Maximum weighted matching, Compare Match, Merge, Model merging}
}
RIS :
TI  - An Approach And A Tool For Merging A Set Of Models In Pairwise Way
AU - M. Boubakir
AU - A. Chaoui
PY - 2021
SN - 0127-9084
JO - Malaysian Journal of Computer Science
VL - 34
IS - 1
SP - 13–33
PB - University of Malaya
AB - Model merging addresses the problem of combining information from a set of models into a single one. This task is considered to be an important step in various software engineering practices. When many (more than two) models need to be merged, the most practiced technique is to perform the merge in a pairwise way, without considering the order of merging. The problem with this technique is that the resulting quality is not guaranteed because it is influenced by such an order. In this paper, we propose a pairwise approach for model merging aiming to provide better results by taking into account the order of merging. This approach proposes to combine models in an iterative process until obtaining only one model. At each iteration, we first compare each pair of models in order to measure the similarity between them and to identify the correspondences between their component elements. This is performed using two heuristic-based operators respectively named compare and match. After that, we identify the most similar pairs of models and merge them using a proposed operator. We have implemented our approach as tool support called 3M and evaluated it on a set of case studies.
KW - Model comparison
KW - Maximum weighted matching
KW - Compare Match
KW - Merge
KW - Model merging
UR - https://ejournal.um.edu.my/index.php/MJCS/article/view/12728
ID - misc-lab-351
ER -