Fig 8 - uploaded by Denis Chêne
Content may be subject to copyright.
Source publication
![Fig. 2. Adaptation time according to McKinley's taxonomy [28].](profile/Denis-Chene/publication/317919397/figure/fig2/AS:614203824025626@1523449011831/Adaptation-time-according-to-McKinleys-taxonomy-28_Q320.jpg)
![Fig. 3. The ISATINE framework for UI adaptation [27].](profile/Denis-Chene/publication/317919397/figure/fig3/AS:614203824042021@1523449011866/The-ISATINE-framework-for-UI-adaptation-27_Q320.jpg)
Context in source publication
Context 1
... user (which is challenging due to latency and lag [13]) and to make this state accurately [19] interpreted by the end-user. -Predictability: refers to the user's ability to predict future system actions for supporting the interaction, the adaptation, or both based on past corresponding actions [19]. Regarding the adaptation predictability (see Fig. 8 for various forms of this property), the user should to some extent predict the be- haviour of the adaptivity algorithm based on past adaptivity actions, which may subsume controllability: the end-user can predict all the better that she is under control of the system thanks to controllability. Accuracy of the prediction posi- tively ...
Similar publications
Citations
... Borrowed from adaptive user interfaces, we believe the following usability criteria are crucial for adaptive mobile maps to be accepted by future users (Dhouib et al., 2017): predictability (users need to understand the conditions of map adaptation and how the map app functions), controllability (users should be able to control the map adaptation process), breadth of experience (the adaptation should limit the available map interface functionalities to simplify the user experience), unobtrusiveness (the map adaptation process should not interrupt the users' main activity), privacy and trust (users should be able to trust the map app and be sure their privacy is protected), transparency (users should be able to understand the map adaptation) (Höök, 2000;Jameson, 2003Jameson, , 2005Jameson, , 2009). In addition to these criteria, Bouzit et al. (2017) propose observability (the map app should make the adaptations perceivable for the user), intelligibility (the map adaptation processes are communicated understandably to the user), intelligibility (could be ensured by different ways, explainability (the adaptation is explained), continuity (the adaptation process is continuously rendered), awareness (the user can perceive how the adaptation is occurring in the map app). ...
Mobile maps are an important tool for mastering modern digital life. In this paper, we outline our perspective on the challenges and opportunities associated with designing adaptive mobile maps that are useful, usable, and accessible to a wide range of users in different contexts. If we claim for adaptive mobile maps to be successful, we need to expand our understanding of map use context, including the physical and digital spaces, user behavior, and individual differences. We identify key challenges, such as the scarcity of knowledge about mobile map use behavior, the need for effective adaptation methods and strategies, user acceptance of adaptive maps, and issues related to control, privacy, trust, and transparency. We finally suggest research opportunities, such as studying mobile map usage, employing AI-based adaptation methods, leveraging the power of visual communication through maps, and ensuring user acceptance through user control and privacy.
... Similarly, Bouzit et al. [Bouzit et al. 2017] formalised a design space for user interface adaptation, alas without modelling the environment. The proposed framework is based on Perception-Decision-Action cycle that is augmented by Learning-Prediction-Action, allowing for UI designs that are descriptive, comparative, and generative. ...
The interaction context (or environment) is key to any HCI task and especially to adaptive user interfaces (AUIs), since it represents the conditions under which users interact with computers. Unfortunately, there are currently no formal representations to model said interaction context. In order to address this gap, we propose a contextual framework for AUIs and illustrate a practical applica- tion using learning management systems as a case study. We also discuss limitations of our framework and offer discussion points about the realisation of truly context-aware AUIs.
... The former refers to the user's ability to adapt to the system's interface. The latter means the system's ability to adapt its interface to the user (Dieterich et al., 1993;Bouzit et al., 2017). The same applies to the HRI domain, where the robot represents the system. ...
Social robots will be part of our future homes. They will assist us in everyday tasks, entertain us, and provide helpful advice. However, the technology still faces challenges that must be overcome to equip the machine with social competencies and make it a socially intelligent and accepted housemate. An essential skill of every social robot is verbal and non-verbal communication. In contrast to voice assistants, smartphones, and smart home technology, which are already part of many people's lives today, social robots have an embodiment that raises expectations towards the machine. Their anthropomorphic or zoomorphic appearance suggests they can communicate naturally with speech, gestures, or facial expressions and understand corresponding human behaviors. In addition, robots also need to consider individual users' preferences: everybody is shaped by their culture, social norms, and life experiences, resulting in different expectations towards communication with a robot. However, robots do not have human intuition - they must be equipped with the corresponding algorithmic solutions to these problems. This thesis investigates the use of reinforcement learning to adapt the robot's verbal and non-verbal communication to the user's needs and preferences. Such non-functional adaptation of the robot's behaviors primarily aims to improve the user experience and the robot's perceived social intelligence. The literature has not yet provided a holistic view of the overall challenge: real-time adaptation requires control over the robot's multimodal behavior generation, an understanding of human feedback, and an algorithmic basis for machine learning. Thus, this thesis develops a conceptual framework for designing real-time non-functional social robot behavior adaptation with reinforcement learning. It provides a higher-level view from the system designer's perspective and guidance from the start to the end. It illustrates the process of modeling, simulating, and evaluating such adaptation processes. Specifically, it guides the integration of human feedback and social signals to equip the machine with social awareness. The conceptual framework is put into practice for several use cases, resulting in technical proofs of concept and research prototypes. They are evaluated in the lab and in in-situ studies. These approaches address typical activities in domestic environments, focussing on the robot's expression of personality, persona, politeness, and humor. Within this scope, the robot adapts its spoken utterances, prosody, and animations based on human explicit or implicit feedback.
... Côté ergonomie, la solution doit amener une augmentation d'efficacité et d'efficience perceptible rapidement par l'utilisateur, conditionnée par son acceptabilité, en particulier vis-à-vis des automatismes déclenchés par la machine. À travers ce paradigme, nous adressons la question cruciale de la bonne répartition du contrôle de l'interface entre l'utilisateur et le système, ceci dans le respect des critères de prédictibilité, d'intelligibilité, de stabilité, de contrôlabilité, de conscientisation et de sécurité [17]. Côté machine, cela implique de concevoir une modélisation suffisamment robuste, souple et réactive, permettant, au travers d'un dialogue étroit et constant entre le système et l'utilisateur de proposer des adaptations pertinentes. ...
As part of the customization of man-machine interfaces, we present in this paper a generic paradigm of dynamic adaptation of the interface to the user’s navigation habits. This involves providing the current user with automated assistance in carrying out his usual tasks, so as to make him more efficient, in particular by reducing his mental load or the number of his actions. This adaptive paradigm, which is not specific to an application, combines in a flexible way: adaptive guidance, adaptive shortcuts, and adaptive automatisms. These different modes of adaptivity are managed by the system during the interaction, in a reliable and reactive manner on the basis of adaptive machine learning leveraging Bayesian theory and dealing with uncertainty in a rigorous way. The architecture of the solution is based on a predictive navigation model underpinned by a finite state machine, and integrating adaptive mechanisms. Our paradigm thus integrates a user activity model, a navigation task model within an interface, and an adaptive interaction model. It offers dynamic navigation assistance by jointly optimizing the nature of the adaptations and the distribution of initiatives between the user and the system, depending on the part of the interface explored and its use. We will endeavor to describe and justify this paradigm from ergonomic and technical angles, supported by a demonstrator running on a touchscreen tablet.
... Finally, Fig. 2 depicts the PDA-LDA cycle employed to structure the UI adaptation according to the theory of control perspective [7]: each entity, the end-user (depicted in blue) or the system (depicted in green) enters a cycle of three stages: the perception (P) of the context before adaptation, the decision (D) to adapt and the action (A) taken in order to adapt. Unlike other frameworks that emphasize the adaptation steps, this cycle acknowledges that both the end-user and the system act symmetrically with these stages, which should be covered to some extent. ...
... Unlike other frameworks that emphasize the adaptation steps, this cycle acknowledges that both the end-user and the system act symmetrically with these stages, which should be covered to some extent. 3 Adaptation granularity (What) 4 User Interface type (What) 5 User interface modality (What) 6 Context Coverage (Why) 7 Adaptation rationale (Why) 6 Context Coverage (Why) 8 Adaptation QAs (What) 12 9 Adaptation location (Where) 10 Adaptation scope level (Where) ...
... If necessary, the transitioner provides the end-user with information on why, how and when the adaptation is performed by requesting the adaptation explainer, which is responsible for explaining and justifying why any adaptation proposal or step will be executed [16]. Finally, an adaptation machine learning system can monitor the whole process over time, learn what the good adaptations are or which are preferred by the end-user, and recommend them in the future [7]. For example, TADAP [27] suggests adaptation operations based on the user's interaction history that the end-user can accept, reject, or re-parameterise by employing Hidden Markov Chains. ...
Adapting the user interface of a software system to the requirements of the context of use continues to be a major challenge, particularly when users become more demanding in terms of adaptation quality. A considerable number of methods have, over the past three decades, provided some form of modelling with which to support user interface adaptation. There is, however, a crucial issue as regards in analysing the concepts, the underlying knowledge, and the user experience afforded by these methods as regards comparing their benefits and shortcomings. These methods are so numerous that positioning a new method in the state of the art is challenging. This paper, therefore, defines a conceptual reference framework for intelligent user interface adaptation containing a set of conceptual adaptation properties that are useful for model-based user interface adaptation. The objective of this set of properties is to understand any method, to compare various methods and to generate new ideas for adaptation. We also analyse the opportunities that machine learning techniques could provide for data processing and analysis in this context, and identify some open challenges in order to guarantee an appropriate user experience for end-users. The relevant literature and our experience in research and industrial collaboration have been used as the basis on which to propose future directions in which these challenges can be addressed.
... Adaptive human-computer interaction is indeed a broad topic in Human-Computer Interaction (HCI) research. Using a software engineering approach, Bouzit et al. proposed the PDA-LPA design space for user interface adaptation [3]. According to these authors, adaptation falls into two categories depending on who is involved in the adaptation process: adaptability refers to the user's ability to adapt the interface, whereas adaptivity refers to the system's ability to perform interface adaptation. ...
Contemporary societies are comprised of individuals very diverse in terms of culture, status, gender and age. In this context, there is no single system behaviour that fits all users, not even considering the traditional “personalization” efforts of adaptable systems, in which individual users can explicitly tailor some system features to their needs. Multicultural and ageing societies demand adaptive interactive systems with the ability to learn about and from their users and adjust their behaviour accordingly. This chapter presents a vision of an intelligent reminder agent to illustrate the current challenges for the design and development of adaptive systems, which are analysed following the what-which-how-then model to cover all aspects of which features to adapt, what to adapt to, when to adapt and how to adapt, with special emphasis on multimodal conversational interaction.
... For each user the system is customized according to their interests and competencies [9,18,33]. A co-evolutionary process between the user and the system is established [4]. However, according to Paternò [34] the main problem in the development of adaptive systems is to structure the information in such a way as to allow customization. ...
To address diverse interaction needs of heterogeneous users' groups, user interfaces must be flexible to accommodate for customization that are specific to each user profile. Although, existing web interfaces provide some flexibility, some problems still remain: a) manual adjustments carried out by end users are required for each web application; b) the flexibility provided by current web interfaces is insufficient to address diverse interaction needs of various users' profiles and c) few users are aware about such options to customize the presentation of web interfaces. To contribute to the customization of user interface according to the needs of diverse users, in this work we asses the suitability of a tool that customize web interfaces based on the needs and preferences of end users. UIFlex is a web-based browser plugin that enables users to define their interaction profile. In this task, users are supported by fifteen web-based design rules that were extracted from the literature and the knowledge of authorities. To customize the presentation of web interfaces, UIFlex relies on a set of rules defined for each individual user and "injects" JavaScript codes, Cascading Style Sheets (CSS) and in some cases HyperText Markup Language (HTML) codes in any page that follows W3C standards. UIFlex was evaluated by 104 users of diverse interaction profiles. The results obtained are promising and suggest that the solution improves the perception that the interactive system performs as desired by users.
... We propose a dynamic adaptation approach that tracks various types of engagement over the course of the learning experience to suggest an adaptation of the game element when learner engagement decreases. To properly describe how this dynamic adaptation approach should work, we can use the PDA/LPA framework proposed by (Bouzit et al., 2017) -see Figure 2. The PDA/LPA framework describes two adaptation cycles on both the user and the system sides. Users perceive an adaptation change, make a decision about this change, and perform an action (PDA). ...
... The system then analyses the new log traces generated by the learner and estimates his/her engagement -Perception -(restart from step 2) This adaptation system raises a few questions that need to be answered: (1) How can we control that these adaptations are effective? ((Bouzit et al., 2017) propose that: "controllability is essential to enable the end-user to be actively involved in any adaptation activity") (2) How can we ensure that these adaptations do not occur too often (creating an unstable environment for the learner), and how can we ensure that adaptations do not occur too rarely (and therefore not reacting quickly enough to losses of engagement)? We provide first answers to these questions within the framework of the LudiMoodle project described in next section. ...
... Finally, we can enrich our dynamic adaptation approach by including the other steps of the PDA/LPA framework cycles (Bouzit et al., 2017). Currently the dynamic adaptation does not make use of the Prediction or Adaptation steps. ...
Network analysis simulations were used to guide decision-makers while configuring instructional spaces on our campus during COVID-19. Course enrollment data were utilized to estimate metrics of student-to-student contact under various instruction mode scenarios. Campus administrators developed recommendations based on these metrics; examples of learning analytics implementation are provided.
... We propose a dynamic adaptation approach that tracks various types of engagement over the course of the learning experience to suggest an adaptation of the game element when learner engagement decreases. To properly describe how this dynamic adaptation approach should work, we can use the PDA/LPA framework proposed by (Bouzit et al., 2017) -see Figure 2. The PDA/LPA framework describes two adaptation cycles on both the user and the system sides. Users perceive an adaptation change, make a decision about this change, and perform an action (PDA). ...
... The system then analyses the new log traces generated by the learner and estimates his/her engagement -Perception -(restart from step 2) This adaptation system raises a few questions that need to be answered: (1) How can we control that these adaptations are effective? ((Bouzit et al., 2017) propose that: "controllability is essential to enable the end-user to be actively involved in any adaptation activity") (2) How can we ensure that these adaptations do not occur too often (creating an unstable environment for the learner), and how can we ensure that adaptations do not occur too rarely (and therefore not reacting quickly enough to losses of engagement)? We provide first answers to these questions within the framework of the LudiMoodle project described in next section. ...
... Finally, we can enrich our dynamic adaptation approach by including the other steps of the PDA/LPA framework cycles (Bouzit et al., 2017). Currently the dynamic adaptation does not make use of the Prediction or Adaptation steps. ...
... If the change brought on by the adaptation is not explained or presented to the learner in a clear and understandable manner this could confuse and could distract the learner from his/her learning activity. In the field of user interface adaptation Bouzit et al. [15] show that change needs to be observable, intelligible, predictable and controllable for the user. We believe therefore that research needs to be done into how these concepts can be applied to educational settings. ...
... Furthermore, as the platform presented multiple quizzes for each maths subject, that were not specifically linked, they felt that the metaphor of the branching tree would resonate well with the learners (this was a way to include the context in the game element design). Figure 12a shows the design tools used, and table 15 shows the full description of the game element designed through the lens of the design space. Generally participants manipulated the cards with ease, however we observed that the participants had difficulties using the "Behaviour Change" dimension as they always selected the same behaviour. ...
... To better explain how our dynamic adaptation system functions I will use the PDA-LPA design space proposed by Bouzit et al. [15] for describing and understanding interface adaptation. The design space is split over two usage loops (one for the end user, and one for the system) that follow two successive cycles: PDA (Perception Decision Action) and LPA (Learning Prediction, Adaptation). ...
Gamification, the use of game elements in non game contexts, is becoming widely used in the educational field to enhance learner engagement, motivation, and performance. Many current approaches propose systems where learners use the same game elements. However, recent studies show that learners react differently to different game elements, and that learner motivation, engagement, and performance can vary greatly depending on individual characteristics such as personality, game preferences, and motivation for the learning activity. Results indicate that in some cases game elements that are not adapted to learners can at best fail to motivate them, and at worst demotivate them. Therefore, adapting game elements to individual learner preferences is important. This thesis was part of the LudiMoodle project, dedicated to the gamification of learning resources to enhance learner engagement and motivation. In this thesis, I propose a new system that adapts relevant game elements to learners using individual characteristics, as well as learner engagement. This work is based on previous results in the general gamification field, as well as more specific results from gamification in education. Our main goal is to propose a generic adaptation engine model, instantiated with specific adaptation rules for our educational context. This manuscript presents four major contributions: (1) A general adaptation engine architecture that can be implemented to propose relevant game elements for learners, using both a static and dynamic adaptation approach; (2) A design space and design tools that allows the creation of relevant and meaningful game elements, in collaboration with the various actors of the gamification process (designers, teachers, learners etc.); (3) A static adaptation approach that uses a compromise between both learners' player profile (i.e. preferences for games) and their initial motivation for the learning task; (4) A dynamic learner model built on a trace-based approach to propose an adaptation intervention when an abnormal decrease in engagement is detected. The adaptation engine was implemented in a prototype for the LudiMoodle project, that was used by 258 learners in 4 different secondary schools in France for learning mathematics. To build this prototype we ran a real world study, where learners used this tool as a part of their normal mathematics course. From this study, we ran multiple analyses to better understand the factors that influence the motivational variations of the learners, and how their interaction traces could predict their engagement with the learning task. These analyses served to evaluate the impact of the adaptation of game elements on learner motivation and engagement, and to build the trace based model used for dynamic adaptation.This work represents a significant advancement for the adaptive gamification field, through a generic model for static and dynamic adaptation, with the former based on individual learner characteristics, and the latter on observed learner engagement. I also provide tools and recommendations for designers, to help explore different game element designs. Finally, I discuss these findings in terms of research perspectives, notably with regards to further possible advancements in the dynamic adaptation domain.
The full text of my thesis can be found the open archive: https://tel.archives-ouvertes.fr/tel-03125624
