Use case diagrams appear early within a UML-based development, structured over the concepts of actors and use cases to capture user requirements of an application. Good modeling practices suggest that use case diagrams should be simple and easy-to-read, two goals that can be achieved by introducing relevant generalizations of actors and use cases. The approach presented in this paper allows, using Formal Concept Analysis and one of its variants, Relational Concept Analysis, to refactor a use case diagram as a whole in order to make it clearer while respecting the semantics of the original diagram. The relevancy of this approach has been confirmed by its implementation as a tool and the results obtained from its application on several representative diagrams.

In the 1980s, the idea of accessing new knowledge by coupling simulation models gave birth to software platforms in agronomy. In most of the current platforms, assembly is achieved using a generic assembly medium (SGA). The question raised in this paper is that of matching a SGA to a software platform. To do this, we propose a functional description framework. This framework combines criteria stemming from software engineering and systems theory. The examples chosen to illustrate the reasoning are those of MODCOM, OpenMI and VLE, for SGAs, and APES and SEAMLESS-IF, for software platforms. In terms of prospects, this work paves the way for automation of the assembly operation using a SGA. MOTS-CLES : modele de simulation, cohesion, couplage, valence, interrelation, ontologie

Dans le cadre de leurs travaux, les chercheurs peuvent etre amenes a assembler des programmes de simulation preexistants. Le mode operatoire repose sur l'hypothese que l'assemblage de programmes, chacun representant un sous-systeme, permet d'acceder a la simulation du systeme global. Cette hypothese n'est pas toujours verifiee et la question se pose de l'identification des sous-systemes manquants. La theorie du systeme general apprehende les systemes sous la forme d'une collection d'actions enchevetrees. L'objet de ce document est de presenter une methode declarative de description de systeme complexe reposant sur cette theorie et d'en montrer une application a un systeme biologique.

Dans le contexte d'un processus de developpement base sur la notation UML, les diagrammes de cas d'utilisation et les modeles ou documents qui les accompagnent pour traduire les besoins utilisateurs contribuent a formaliser la definition des besoins. Ils guident egalement toutes les etapes de construction du systeme depuis l'etablissement du cahier des charges jusqu'a la realisation des dossiers de test. Nous etudions la faisabilite d'une approche permettant la correction de certains defauts des diagrammes de cas d'utilisation UML (lies a une mauvaise utilisation des concepteurs) par introduction d'acteurs et de cas plus generaux, ainsi que de relations de specialisation/generalisation appropriees. Les differents cas de refactorisation par introduction de generalisations sont consideres de maniere systematique, en utilisant la structure du diagramme et une segmentation des noms des cas. Cette etude debouche sur une utilisation de l'Analyse Formelle de Concepts pour traiter la refactorisation de maniere globale. Une chaine d'outils permet de partir d'un diagramme de cas d'utilisation UML et d'analyser differentes generalisations possibles qui aboutissent le plus souvent a une simplification du diagramme.

In Model Driven Engineering (MDE), model transformations are basic and primordial entities, thus easing their design and implementation is an important issue. A quite recently proposed way to create model transformations consists in deducing a transformation from examples of transformed models. Examples are easier to write than a transformation program and are often already available. We propose in this paper a method based on a machine learning method of the lattice domain, the Relational Concept Analysis, and an implementation of this method.

Since the early seventies, numerous numerical programs have been developed to simulate biophysical processes. The current challenge facing scientific communities is to access any program using their usual language. Migration is a way of dealing with the challenge. The conventional methodology to build up migration functions is the rule-based method. An alternative consists in using the category theory. This paper illustrates the methodology applied to bidirectional migration of a program written using a language based on the system theory (SIMILE) to a biological modelling language (APES EU SEAMLESS integrated project) constructed using design patterns. Using the category theory enabled us (i) to provide a mathematical formalism for composition of the program features, i.e. the architecture and semantic function, (ii) to identify aspects which could not be preserved in the migration because of language expressiveness, and (iii) to automate the migration.

Current processor and multiprocessor architectures are almost all based on the Von Neumann paradigm. Based on this paradigm, one can build a general-purpose computer using very few transistors, e.g., 2250 transistors in the first Intel 4004 microprocessor. In other terms, the notion that on-chip space is a scarce resource is at the root of this paradigm which trades on-chip space for program execution time. Today, technology considerably relaxed this space constraint. Still, few research works question this paradigm as the most adequate basis for high-performance computers, even though the paradigm was not initially designed to scale with technology and space.In this article, we propose a different computing model, defining both an architecture and a language, that is intrinsically designed to exploit space ; we then investigate the implementation issues of a computer based on this model, and we provide simulation results for small programs and a simplified architecture as a first proof of concept. Through this model, we also want to outline that revisiting some of the principles of today's computing paradigm has the potential of overcoming major limitations of current architectures.

Le travail concerne l'assemblage de sous-systemes biologiques pour acceder a une representation fonctionnelle du systeme global, dans le cas d'interactions biologiques complexes (symbiose, etc.). La methode proposee utilise l'enonce de connaissances expertes decrites de facon litterale pour etablir le programme d'interfacage des sous-systemes. L'utilisation de pronoms et adverbes interrogatifs confere un cadre formel a la production de graphes a partir des descriptions litterales. En matiere de resultat, la methode permet (i) d'acceder au vocabulaire du programme d'assemblage, (ii) d'identifier les fonctions de transformation a mettre en place, et (iii) livre la valence requise par les programmes pour assurer l'assemblage. D'un point de vue operationnel, la demarche generique proposee est une construction theorique qui reste a valider dans des cas concrets d'application. ABSTRACT. The work concerned the assembly of biological sub-systems to gain access to a functional representation of the overall system, in the case of complex biological interactions (symbiosis, etc.). The proposed method uses the statement of expert knowledge described in a literal way to establish the sub-system interface program. The use of interrogative pronouns and adverbs provides a formal framework to produce graphs from literal descriptions. In terms of results, the method enables (i) access to the vocabulary of the assembly program, (ii) identification of the transformation functions to be established, and (iii) an estimation of the valence required by programs to ensure assembly. From an operational perspective, the generic approach proposed is a theoretical construct that remains to be validated in concrete cases of application.(Resume d'auteur)

In the systemic approach, the system is perceived as an action or a collection of overlapping actions expressed in reference to Time, Space, and Morphology (or Energy). When the system is studied by different disciplines, the referentials differ, as well as the semantics of terms used to describe the action. In order to establish the vocabulary of a collection of actions involving several disciplines, we propose a formal method for describing each action. The linguistic-based method enables (i) transcription of the literal description of an action in a semantic network, and (ii) building of a vocabulary in a formal setting. The method is illustrated through a complex biological system, i.e. the mutualistic relationship between two vine pests, while focusing particularly on temporality. The method provides a support for implementing multidisciplinary around a complex system. (Resume d'auteur)