Décision multicritère à base de traces pour les applications interactives à exécution adaptative

Abstract : Our work deals with software architectures for adaptive interactive applications. We assume that one application is structured with contextual interaction sequences called situations. Users perform actions in successive situations to reach one or more predefined designer’s objectives. During its execution, it could happen that the user cannot fulfil designer’s logic because of some system’s blockings or missing data. Our challenge is to propose a method that chooses the most appropriate situation according to the given one. We propose to improve the decision-making process by using the generated traces during previous executions. These traces represent users’ interactions and system activity logs. A trace-based system collects and manages all users’ generated traces (logs). Our main contributions are: the design of a trace-based algorithm for criteria weighting; the design of an alternatives determination algorithm; the design and the formalisation of the users’ choice (using the trace-based subjective logic) and the system’s choice (using the trace-based PROMETHEE II) to classify all the identified alternatives and the aggregation of different choices to suggest to the user the right option to follow. A Tamagotchi case study is presented to validate our contributions.
Document type :
Complete list of metadatas

Cited literature [93 references]  Display  Hide  Download

Contributor : Abes Star <>
Submitted on : Wednesday, September 28, 2016 - 5:23:09 PM
Last modification on : Thursday, May 17, 2018 - 3:59:37 AM


Version validated by the jury (STAR)


  • HAL Id : tel-01256601, version 2



Hoang Nam Ho. Décision multicritère à base de traces pour les applications interactives à exécution adaptative. Interface homme-machine [cs.HC]. Université de La Rochelle, 2015. Français. ⟨NNT : 2015LAROS024⟩. ⟨tel-01256601v2⟩



Record views


Files downloads