糖心Volg

Denis Lalanne

photo

 denis.lalanne@unifr.ch
 +41 26 300 8472
 

Professeur路e ordinaire
D茅partement d'informatique

PER 21 bu. A429
Bd de P茅rolles 90
1700 Fribourg
PER 21, A429
+41 26 300 8472

Biographie

Denis Lalanne est directeur de l'institut et professeur ordinaire au département d'informatique de l'université de Fribourg. Il dirige le groupe « human-building interaction » au sein du et représente la Suisse à l'IFIP TC13. Après un doctorat à l'École polytechnique fédérale de Lausanne (EPFL) sur la créativité assistée par ordinateur, un post-doc dans le groupe USER au centre de recherche IBM Almaden, une année d'enseignement et de recherche à l'Université d'Avignon, et une expérience d'ingénieur en utilisabilité dans une start-up suisse, il a rejoint l'Université de Fribourg en 2002 pour travailler sur le PRN IM2 du FNS en tant que chercheur senior puis il est devenu professeur en 2015 sur le thème de l'interaction humain-machine. Il fonde l'institut Human-IST en 2017 et l'a développé depuis. Il s'intéresse à différents aspects du domaine de l'interaction humain-machine, avec un accent sur l'interaction multimodale et la visualisation de l'information. Plus récemment, ses recherches se sont concentrées sur le thème de l'interaction humain-bâtiment pour comprendre et améliorer le confort des occupants des bâtiments et sur le thème de l'interaction humain-IA pour améliorer/évaluer le travail en équipe ou pour soutenir l'accessibilité/la réadaptation.

Recherche et publications

  • Liste des publications
    184 publications


    Maximiliano Jeanneret Medina, Yong-Joon Thoo, C茅dric Baudet, Jon E. Froehlich, Nicolas Ruffieux, Denis Lalanne, ACM Transactions on Accessible Computing (2025) | Article

  • Projets de recherche

    Tour Data Visualytics
    Statut: En cours

    顿茅产耻迟 01.03.2024
    Fin 28.02.2026
    Financement Innovation 

    Swiss Post is currently deploying new algorithms to optimize their delivery routes, but to further improve their efficiency, it is necessary to leverage the tacit knowledge of their delivery personnel. According to statistics collected on the field, 10% of all routes are currently not performed according to the planification.

    The aim of this project is to gather and utilize the tacit knowledge of delivery personnel to optimize the delivery routes of postal companies. Formalizing the tacit knowledge is hard, even though it is the leading cause for deviation patterns planned and realized tours. Using learning-based approaches to analyze patterns of deviation will lead to more sustainable last-mile delivery operations, more satisfied drivers and higher service quality.

    The project will focus on three main topics (1) detect the deviations between planned and realized tour, (2) classify them with reinforcement learning algorithm or by emitting a labelling query to the postman, (3) process and visualize the augmented data.

    This approach creates a cycle through an indirect dialogue between the postmen and the Post Planner algorithm. The augmented data will be processed to generate visual analytics at micro and macro-levels (from distribution base to whole country). Base managers will use these analytic tools to better monitor the efficiency of the planning algorithm and to support dialogue with postmen to improve internal processes. This augmented data can also be leveraged by next-generation planners to further optimize delivery routes. 

    Swiss Post Human-IST Lab
    Statut: En cours

    顿茅产耻迟 01.01.2024
    Fin 31.12.2027
    Financement Innovation 

    Understanding ALGorithmic gatekeepers to promote EPIstemic welfare (ALGEPI)
    Statut: En cours

    顿茅产耻迟 01.05.2023
    Fin 30.04.2027
    Financement FNS 
    Technological and economic developments have led to the availability of an overwhelming quantity of digital content. Therefore, it has become crucial, in particular for media content providers, to incorporate algorithmic gatekeepers, which filter, rank and recommend content. Are these algorithmic gatekeepers undermining media鈥檚 contribution to epistemic welfare? In ALGEPI, we start from the novel concept of epistemic welfare, defined as the individuals鈥� right to know and be exposed to trustworthy, independent and diverse information while respecting individual rights to their own data. By connecting legal, political, technological and sociocultural perspectives, we will develop a conceptual framework for epistemic welfare. This framework will allow us to understand the effects of algorithmic gatekeepers on epistemic welfare. The new interdisciplinary research program will build on multi-method empirical research applied to algorithmic gatekeepers. Thanks to the theoretical and empirical findings, we will be able to develop normative instruments to align algorithmic gatekeeping with epistemic welfare. Taking the novel concept of epistemic welfare as a starting point allows all project partners to embark on a joint, integrated exploration of the concept. This may lead to a paradigmatic shift in the conceptualization of the impact of algorithmic gatekeepers in media sectors, proposing to expand the notions of media pluralism and consumer welfare to epistemic welfare.

    SWICE (Sustainable Well-being for the Individual and the Collectivity in the Energy transition )
    Statut: En cours

    顿茅产耻迟 01.01.2023
    Fin 31.12.2031
    Financement Autre 

    The SWICE project aims to identify and quantify the energy saving potential and opportunities for increased quality of life that can emerge from future urban scenarios involving new modalities of living and working, changes in mobility behaviours, and different business models.

    Financed by the  (Call 1-2021 “Living and Working)

    iKnowU
    Statut: Termin茅

    顿茅产耻迟 01.01.2020
    Fin 31.12.2023
    Financement Innovation 

    Preventing 鈥淐omfort Degradation鈥� in Workplaces with Logi Personal Companion
    Statut: Termin茅

    顿茅产耻迟 01.09.2018
    Fin 01.05.2021
    Financement Innovation 

    28猫me 颁辞苍蹿茅谤别苍肠别 Francophone sur l鈥橧nteraction Homme-Machine
    Statut: Termin茅

    顿茅产耻迟 01.10.2016
    Fin 31.12.2016
    Financement FNS 
    Pour sa 28猫me 茅dition, la conf茅rence francophone sur l'Interaction Homme-Machine vient en Suisse, 脿 Fribourg. L'茅v猫nement co茂ncidera aussi avec les 20 ans de l'Association Francophone d'Interaction Homme-Machine (AFIHM). L鈥橝ssociation Francophone d鈥橧nteraction Homme-Machine et l鈥櫛跃北贡鸢蹙背倜� de Fribourg, avec le soutien de la HES-SO, sont heureux de vous accueillir 脿 IHM鈥�16, la 28猫me 颁辞苍蹿茅谤别苍肠别 Francophone sur l鈥橧nteraction Homme-Machine. La conf茅rence aura lieu du 25 au 28 octobre 2016, sa pr茅sidence 茅tant assur茅e par Denis Lalanne (鲍苍颈惫别谤蝉颈迟茅 de Fribourg) et Laurence Nigay (鲍苍颈惫别谤蝉颈迟茅 Grenoble Alpes). IHM16 sera, nous l鈥檈sp茅rons, un lieu d'茅changes entre les diff茅rents acteurs de la recherche en Interaction Homme-Machine. Nos ambitions avec cette 茅dition d'IHM sont de : - Encourager la participation de jeunes chercheurs - Augmenter les liens entre laboratoires et entreprises, notamment locales, autour des probl茅matiques d鈥橧HM - Favoriser la pluridisciplinarit茅 (informatique, psychologie, sociologie) - Promouvoir le domaine de l鈥橧nteraction Homme-Machine en Suisse et de tisser un lien fort avec le reste de la communaut茅 en Interaction Homme-Machine - Profiter de cet 茅v茅nement pour sensibiliser le canton et ses institutions de l鈥檌mportance de cette discipline

    EmotiBoard: Real-time Emotion Detection and Visualization
    Statut: Termin茅

    顿茅产耻迟 01.09.2011
    Fin 31.10.2012
    Financement Autre 

    Humanitics - Visualizing Migration Flows and their Development in Time
    Statut: Termin茅

    顿茅产耻迟 01.04.2011
    Fin 31.05.2013
    Financement FNS 
    Humanitics - Visualizing Migration Flows and their Development in Time

    Human centered design and evaluation
    Statut: Termin茅

    顿茅产耻迟 01.01.2010
    Fin 31.12.2013
    Financement FNS 
    The goal of IP2 in the third phase of IM2 is to validate and generalize the technologies developed through research in the first two phases of IM2. The capacity to generalize these technologies will be demonstrated (1) through their application to new environments, other than smart meeting rooms, in combination with third-party technology, and (2) through their acceptance by various groups of users, which will be shown by formal studies of user requirements and user-oriented evaluation. The environments that are planned to be designed and implemented in IP2 will follow a novel approach which departs from the smart meeting rooms of Phases 1 and 2. The new instrumented environments are intended to augment teamwork by using 鈥渓ightweight鈥� devices which will support team processes without being in the foreground, and without requiring from users a series of complex steps to initialize and run them. Neither participants, nor external operators will need to devote their attention to the tools, but instead will be helped to focus on their meeting purpose. Technology will thus be in the background, embedded into the physical environment of the rooms, that is, into furniture such as tables, lamps, or walls. The approach of IP2 believes that intelligence is distributed among meeting participants or individual users, and does not lie directly in the technology, which simply provides hints and scaffolds productive interactions. Therefore, IP2 has two related objectives: - Develop new, lightweight applications, making use of IM2 technologies, and mainly oriented towards teamwork support. - Develop and apply formal user-centric evaluation methods, targeting the evaluation of a selection of systems, both from phase 1 and 2 technology, and from the new applications.

    Humanitics: Visual Analytics for All
    Statut: Termin茅

    顿茅产耻迟 01.04.2009
    Fin 31.03.2011
    Financement FNS 
    The overall goal of this project is to develop visual analytics tools for public organizations to enhance knowledge discovery, analytical capabilities, decision making, information exchange and communication. The targeted users are from countries all around the world, including least developed countries that generally do not have access to this type of technology. Each year, the United Nation manipulates and reports numerous data and information concerning world-wide health, illicit drug trade, environment and global climate change, diseases, energy, conflict, and humanitarian development concerns. The UN urgently needs authoring tools to analyse these data and cross correlate them in a visual way, not only to improve visual understanding, but also for analysts to produce meaningful visualizations to present their findings. UN also needs user friendly tools to browse through these findings and produce reports; tools that should be usable by everybody (all countries). Visual analytics is the science of analytical reasoning facilitated by interactive visual interfaces. People use visual analytics tools and techniques to synthesize information and derive insight from massive, dynamic, ambiguous, and often conflicting data; detect the expected and discover the unexpected; provide timely, defensible, and understandable assessments; and communicate assessment effectively for action. The US has already started a NSF program on visual analytics, driven principally by security issues, to track terrorism and abnormal behaviours. This National Science Foundation (NSF) program is a long run initiative and implicates about 20 American institutions (academics and industrials). 鈥淭he U.S. Government faces a critical challenge in identifying and preventing attacks on U.S. soil. The National Visualization and Analytics Center (NVAC) is a national and international resource providing strategic leadership and coordination for visual analytics technology and tools. NVAC supports the Department of Homeland Security鈥檚 mission to secure our homeland and protect the American people by giving analysts and emergency responders technology and capabilities to: Detect, prevent, and reduce the threat of terrorist attacks; Identify and assess threats and vulnerabilities to our homeland; Recover and minimize damage from terrorist attacks, should they occur. 鈥� We believe this multi-billon dollar (USD) American NSF initiative on visual analytics should be complemented by an European initiative using visual analytics not for tackling security and terrorism issues but driven by humanistic and world-wide needs and issues. Information access and understanding is crucial in our modern society, not only to insure security and control of rich countries, but also for all countries to understand the impact of their public policies on a world-wide perspective. Information is power and everybody, including third parties countries, should have access to it. The associated research challenges cover (1) setting up a toolkit, integrating state of the art visualization and mining techniques, facilitating Visual Analytics application development by endusers and (2) developing mechanisms for analysts to annotate data and visualizations, in order to build collectively and incrementally their knowledge.

    VisMaster CA: Visual Analytics - Mastering the Information Age
    Statut: Termin茅

    顿茅产耻迟 01.08.2008
    Fin 30.09.2010
    Financement Europe 
    One of the most important challenges of the emerging Information Age is to effectively utilise the immense wealth of information and data acquired, computed and stored by modern information systems. On the one hand, the appropriate use of available information volumes offers large potential to realize technological progress and business success. On the other hand, there exists the severe danger that users and analysts easily get lost in irrelevant, or inappropriately processed or presented information, a problem which is generally called the information overload problem. Visual Analytics is an emerging research discipline developing technology to make the best possible use of huge information loads in a wide variety of applications. The basic idea is to appropriately combine the strengths of intelligent automatic data analysis with the visual perception and analysis capabilities of the human user.We propose a Coordination Action to join European academic and industrial RandD excellence from several individual disciplines, forming a strong Visual Analytics research community. An array of thematic working groups set up by this consortium will focus on advancing the state of the art in Visual Analytics. Specifically, the working groups will join excellence in the fields of data management, data analysis, spatial-temporal data, and human visual perception research with the wider visualisation research community. This Coordination Action will (1.) form and shape a strong European Visual Analytics community, (2.) define the European Visual Analytics Research Roadmap, (3.) expose public and private stakeholders to Visual Analytics technology and (4.) set the stage for larger follow-up Visual Analytics research initiatives in Europe.

    The Risk Manager Dashboard
    Statut: Termin茅

    顿茅产耻迟 01.01.2008
    Fin 31.12.2008
    Financement Autre 
    Visualization for assessing and managing risk in corporate networks through metrics and user behavior analysis. Summary This project involves the Department of Computer Science of the 糖心Volg (DIUF) and NEXThink S.A., a start-up software company developing a new generation of security management solutions. The project will develop novel visualizations for risk monitoring in network security. New visual metrics will be developed in a way that risk can be correctly captured and easily understood by non-technical users. The project will also develop interactive visualizations to analyze the causes and consequences of changes in risk levels. The new system will be used to expand deployment of the solution with early-adopter customers and support large to very-large organizations with several dozen of thousands of employees.

    Human Computer Interface (HMI)
    Statut: Termin茅

    顿茅产耻迟 01.01.2006
    Fin 31.12.2009
    Financement Autre 
    Imagine that you missed a two-hour meeting with your colleagues, but that this meeting was recorded. You want to know what you missed, but you do not want to replay the entire meeting. Instead, you want to find quickly just the parts that most interest you.While other researchers within IM2 develop analysis and recognition technologies to address this need, the primary objective of this work is to build on these other technologies to develop interactive browsing systems.The objectives of this project fall into three broad categories: 1. Design novel interactive meeting browsers. 2. Develop working prototypes, suitable for human testing. 3. Evaluate the usability of these interactive prototypes with human subjects. In particular, the 糖心Volg will design three examples of document-centric multimedia meeting browsers: FriDoc and FaericWorld. A user-centred design approach will include use-case definition, mock-ups and low-fidelity prototype creation. 鈥� FriDoc, i.e. FRIbourg DOCument-centric multimedia meeting browser, will be improved to integrate most of the available document annotations and document multimodal alignments. Further, FriDoc will be replicated using JFerret. 鈥� FaericWorld builds upon FriDoc to navigate across collections of multimedia documents.The system will explore the automatic creation and use of multimodal links based on time, words, people, geographic locations, and references (citations, hyperlinks, etc.). FaericWorld will browse through collections of audio/video meeting recordings, conferences, TV news (with text captions), and documents such as newspapers.

    MeModules
    Statut: Termin茅

    顿茅产耻迟 01.01.2006
    Fin 31.12.2009
    Financement Autre  Voir la fiche du projet
    Tangible interfaces can be useful to establish links between our memory and information. We call this type of tangible interface "MEMODULES". They roost abstract information in the real world through tangible reminders. MEMODULES are tiny tagged physical objects containing a link towards information sources. We consider them as tangible hypermarks, i.e. physical object embedding a hyperlink to abstract information.

    Network Security Intelligence through Distributed User Behaviour Modelling and Interactive Visualizations
    Statut: Termin茅

    顿茅产耻迟 01.01.2006
    Fin 30.04.2007
    Financement Autre 
    This project involves the Department of Computer Science of the 糖心Volg (DIUF) and NEXThink S.A., a start-up software company developing a new generation of security management solutions. The project will develop distributed user behaviour modelling and leading-edge visualizations for NEXThink Security Intelligence Management Solution. The new system will be used to expand deployment of the solution with early-adopter customers and support large to very-large organizations with several dozen of thousands of employees. Our goal on this project is to combine two approaches: AI driven and User centric. First, a novel AI approach that address and removes the issues mentioned above through the use of an untapped dimension: user and networked application behaviour to modelize the activity generated on a computer network. Secondly, these data mining methods are combined with information visualization techniques in order to filter and present to the administrator the relevant information; data-mining researchers believe that statistical algorithms and machine learning can be relied on to find interesting patterns, while information visualization researchers believes in the importance of letting users' visual systems lead them to hypothesis making. Our goal is to combine both approaches.

    Human-machine integration
    Statut: Termin茅

    顿茅产耻迟 01.01.2005
    Fin 31.12.2009
    Financement FNS 
    This IP has a research component (pursuing the most promising and/or fundamental research directions initiated in IM2-II), as well as a strong integration and evaluation component. Hence, besides further pursuing some of the most promising research direction in multimodal processing in the strict context of the IM2 vision, one of the objectives of IP1 will be to extend the applications of multimodal technologies, within the human meeting and conference framework, towards more integrated systems that work in real time, with human intervention only when required. Two of the recent prototypes developed in IM2, the Automatic Content Linking Device and the Mobile Meeting Assistant appear as particularly promising starting points for system integration in IM2 Phase III.

    Document Integration (DI)
    Statut: Termin茅

    顿茅产耻迟 01.01.2002
    Fin 31.12.2005
    Financement Autre 
    The National Center of Competence in Research (NCCR) on Interactive Multimodal Information Management, in brief IM2, is aimed at the advancement of research, and the development of prototypes, in the field of man-machine interaction. The NCCR is particularly concerned with technologies coordinating natural input and output modes (such as speech, image, pen, touch, gestures, etc.). The main objective of IM2 Document Integration project is to align various types of documents with video and speech recordings of meetings. Different kind of static documents used during a meeting, either distributed in paper form or projected on a screen, are analyzed and compared to video and speech data in order to allow further linking between the different modalities. In other words, the main goal of the project is to bridge the gap between non-temporal documents and other temporal medias.

    IM2.DI
    Statut: Termin茅

    顿茅产耻迟 01.01.2002
    Fin 31.12.2005
    Financement FNS 

Enseignement et cours

Afficher dans le programme des cours