Skip to Main content Skip to Navigation

Contribution à une méthode de conception et génération d'interface homme-machine plastique

Abstract : Doors to new prospects of using interactive systems have been opened, thanks to the evolution on all the levels of the communication means, and to technological progress leading in particular to new mobile terminal generation. Pervasive computing lets consider a new interactive system generation, and requires new methods of Human-Computer Interaction. From now on the interactive systems must adapt to the context of use, in order to preserve their usability, without need for expensive redesign and reimplementation. Research works relate to new types of UI (User-Interface) which are known as sensitive to context, in other word plastic, with various capacities of integration of the plasticity notion. However, in the majority of methods, the adaptation is statically predefined by the designer at design time. Then, at the time of a need for adaptation following a contextual change, the UI must be transferred at design phase. Moreover, the possibility of evaluating the adaptation quality at runtime misses often. We are interested, as part of the thesis, in the UI having the capacity to adapt dynamically to the context of use, by taking into account the contextual changes without any need to return to design phase. In this research field, and starting from the plastic UI concept, our contribution consists in generating such UI starting from an abstract UI model specified already in an interactive systems specification and design method and/or a task model. Our method is based on the notion of design patterns; those are used during the passage to the concrete UI and during the adaptation. The system architecture is supported by a composition based on business components; those have the capacity to change their presentation facet dynamically. This principle is adopted as a solution for the dynamic adaptation to the context of use. Our method is also based on the notion of machine learning. The integration of a technique of machine learning allows continuing to develop the knowledge base of the system at runtime, in order to preserve the utility of the adaptation. The suggested method is illustrated on two cases of study: the first relates to a nomadic application of guidance touristic. Second is borrowed from the industrial supervision domain.
Document type :
Complete list of metadatas
Contributor : Julie Cagniard <>
Submitted on : Thursday, December 10, 2020 - 12:07:40 PM
Last modification on : Tuesday, December 15, 2020 - 3:31:58 AM


Files produced by the author(s)


  • HAL Id : tel-03015788, version 1



Mohamad-Anas Hariri. Contribution à une méthode de conception et génération d'interface homme-machine plastique. Informatique [cs]. Université de Valenciennes et du Hainaut-Crambrésis, 2008. Français. ⟨NNT : 2008VALE0018⟩. ⟨tel-03015788⟩



Record views


Files downloads