Towards Automatically Improving Package Structure While Respecting Original Design Decisions

Abstract : Recently, there has been an important progress in applying search-based optimization techniques to the problem of software re-modularization. Yet, a major part of the existing body of work addresses the problem of modularizing software systems from scratch, regardless of the existing packages structure. This paper presents a novel multi-objective optimization approach for improving existing packages structure. The optimization ap- proach aims at increasing the cohesion and reducing the coupling and cyclic connectivity of packages, by modifying as less as pos- sible the existing packages organization. Moreover, maintainers can specify several constraints to guide the optimization process with regard to extra design factors. To this contribution, we use the Non-Dominated Sorting Genetic Algorithm (NSGA-II). We evaluate the optimization approach through an experiment covering four real-world software systems. The results promise the effectiveness of our optimization approach for improving existing packages structure by doing very small modifications.
Type de document :
Communication dans un congrès
Working Conference on Reverse Engineering, Sep 2013, Glasgow, United Kingdom. 2013
Liste complète des métadonnées

Littérature citée [19 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00862063
Contributeur : Lse Lse <>
Soumis le : dimanche 15 septembre 2013 - 21:46:42
Dernière modification le : jeudi 11 janvier 2018 - 06:22:25
Document(s) archivé(s) le : jeudi 6 avril 2017 - 20:40:03

Fichier

Abde13a-ImprovingPackageStruct...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00862063, version 1

Citation

Hani Abdeen, Houari Sahraoui, Osama Shata, Nicolas Anquetil, Stéphane Ducasse. Towards Automatically Improving Package Structure While Respecting Original Design Decisions. Working Conference on Reverse Engineering, Sep 2013, Glasgow, United Kingdom. 2013. 〈hal-00862063〉

Partager

Métriques

Consultations de la notice

321

Téléchargements de fichiers

211