Antonin Danalet

New article in Journal of Choice Modelling

Our article "Location choice with longitudinal WiFi data" is now online, published in the Journal of Choice Modelling.

Location choice with longitudinal WiFi data
Danalet, A., Tinguely, L., de Lapparent, M., and Bierlaire, M.
Location-aware data collection technologies provide new insights about location choices. Only a few dynamic models of location choice exist in scientific literature. To our knowledge, none of them correct for serial correlation. In this paper, we model choice of catering locations on a campus using WiFi traces. We use the [Wooldridge, J. M. (2005). Simple solutions to the initial conditions problem in dynamic, nonlinear panel data models with unobserved heterogeneity, Journal of Applied Econometrics 20(1): 39-54.] correction method that deals with the initial values problem and related endogeneity bias in estimation. Cross-validation, price elasticity and simulation of a scenario predicting the opening of a new catering location are presented. Predicted market shares of the new catering location correspond to point-of-sale data of the first week of opening.
Danalet, A., Tinguely, L., de Lapparent, M., and Bierlaire, M. (2016). Location choice with longitudinal WiFi data, Journal of Choice Modelling 18:1-17. doi:10.1016/j.jocm.2016.04.003
[Bibtex]

You can read it by following the doi:10.1016/j.jocm.2016.04.003 if you have access to this journal, or follow this link for direct, free access for 50 days from today.

Posted by Antonin Danalet at 13:45
Comments (0)
La participation à l'échelle de l'auditoire

Le Monde publie aujourd'hui un article sur l'utilisation de Twitter dans les universités françaises :

En amphi de droit, Twitter remplace la main levée

Lire l'article

L'article présente l'utilisation de Twitter dans les auditoires, comme un moyen pour les étudiants d'entrer en contact avec les enseignants.

J'aimerais ajouter un récit d'une expérience qui a été faite à l'EPFL ci-dessous.

Un système de vote interactif

A l'EPFL, dans le cadre du cours "Enjeux mondiaux - Mobilité" destiné aux étudiants de 1e année Bachelor, un système de vote interactif ("clickers") permettait aux étudiants d'exprimer leurs préférences à propos de diverses affirmations. Les étudiants prennent des décisions quotidiennement au sujet de leur mobilité. Le système de vote interactif imitait leurs décisions quotidiennes et politiques. Les résultats étaient projetés, commentés et débattus en classe. De nombreux étudiants volontaires ont expliqué leur vote au micro (nécessaire, avec 300 étudiants inscrits lors du premier cours). Les étudiants ont à nouveau voté sur chaque question à la fin du semestre. Le but était d'observer l'évolution de leur opinion entre leurs croyances a priori et leur opinion informée après le cours.

L'exemple du paradoxe de Braess

La première affirmation du semestre, avant un cours sur le paradoxe de Braess, était

La construction de routes additionnelles fluidifie le trafic.

62.6% des étudiants étaient d'accord ou totalement d'accord avec cette affirmation, alors que 26.3% n'étaient pas d'accord ou pas du tout d'accord. A la fin du semestre, 22.0% des mêmes étudiants étaient d'accord ou tout à fait d'accord et 62.1% n'approuvaient pas ou pas du tout cette affirmation.

Acceptation du système par les étudiants

Voter sur leurs décisions quotidiennes ou politiques a été évalué positivement par plus de 85% des étudiants. Plus de 65% des étudiants se sont déclarés très satisfaits du système.

J'en profite pour rappeler que ce cours a été documenté dans un article de l'International Journal of Engineering Education :

Holzer, A., Vonèche Cardia, I., Bendahan, S., Berne, A., Bragazza, L., Danalet, A., Fasoli, A., Feige, J. N. N., Gillet, D., Isaac, S., Le Duc, I., Preissmann, D., and Tormey, R. (2016). Increasing the Perspectives of Engineering Undergraduates on Societal Issues through an Interdisciplinary Program, International Journal of Engineering Education 32(2(A)):1-11.

La notice bibliographique se trouve sur Infoscience et sur ma page perso sur le site du Laboratoire Transport et Mobilité. Pour ceux qui n'ont pas la chance d'avoir accès à ce journal, une copie de l'article peut être téléchargée sur ResearchGate.

Posted by Antonin Danalet at 14:51
Comments (1)
Location choice with longitudinal WiFi data, paper accepted

Our paper entitled "Location choice with longitudinal WiFi data" has been accepted for publication in the Journal of Choice Modelling today.

Location choice with longitudinal WiFi data
Danalet, A., Tinguely, L., de Lapparent, M., and Bierlaire, M.
Location-aware data collection technologies provide new insights about location choices. Only a few dynamic models of location choice exist in scientific literature. To our knowledge, none of them correct for serial correlation. In this paper, we model choice of catering locations on a campus using WiFi traces. We use the [Wooldridge, J. M. (2005). Simple solutions to the initial conditions problem in dynamic, nonlinear panel data models with unobserved heterogeneity, Journal of Applied Econometrics 20(1): 39-54.] correction method that deals with the initial values problem and related endogeneity bias in estimation. Cross-validation, price elasticity and simulation of a scenario predicting the opening of a new catering location are presented. Predicted market shares of the new catering location correspond to point-of-sale data of the first week of opening.

The preprint (i.e., prior to peer-review) of this paper can be found on the TRANSP-OR website and on Infoscience, EPFL institutional archive. The research data associated to this paper can be found on Infoscience and on Zenodo. The peer-reviewed paper is not available yet.

Posted by Antonin Danalet at 15:31
Comments (1)
A "light" version of my thesis now available

My thesis is now available as a 2Mo-PDF file on Infoscience. The full PDF file is still available (22Mo). The new light version will help referencing it. For Google scholar for example, "each file must not exceed 5MB in size".

The images with full resolutions can be found in the large PDF and on Flickr.

Posted by Antonin Danalet at 12:10
Comments (0)
Le wifi est le moyen le plus facile pour vous traquer

J'étais interviewé ce matin sur la La 1ère, dans l'émission On en parle, par Henry Buxant et Philippe Girard (qui ont presque la même voix, je tiens à le préciser :).

Le wifi est le moyen le plus facile pour vous traquer
Chaque téléphone a son propre identifiant qui permet de le localiser via le wifi notamment. Une manière d’étudier précisément les déplacements d’une personne.

"On en parle" reçoit Antonin Danalet, chercheur à l’EPFL qui a étudié les déplacements de 2'000 personnes sur les campus de l’Ecole polytechnique lausannoise. Un travail éloquent qui montre le potentiel et les dangers des objets connectés. Par Henry Buxant.
Ecouter l'interview
Posted by Antonin Danalet at 11:30
Comments (0)
New paper on Global issues course at EPFL

We have just published an article in the International Journal of Engineering Education about the Global issues program at EPFL, entitled Increasing the Perspectives of Engineering Undergraduates on Societal Issues through an Interdisciplinary Program (click here to access the article). It is about a new course teaching the interplay between societal challenges and technical solutions to EPFL first year students.

Increasing the Perspectives of Engineering Undergraduates on Societal Issues through an Interdisciplinary Program
Holzer, A., Vonèche Cardia, I., Bendahan, S., Berne, A., Bragazza, L., Danalet, A., Fasoli, A., Feige, J. N. N., Gillet, D., Isaac, S., Le Duc, I., Preissmann, D., and Tormey, R.
To tackle the challenges of the 21st century, future scientists and engineers have to understand the interplay between societal challenges and technical solutions as early as possible in their education. They also have to develop the communication and the teamwork skills required to be effective professionals. To address this issue, the Ecole Polytechnique Fédérale de Lausanne (EPFL) introduced a new Global Issues program to all 1800 first year engineering students. In this paper, we present this novel program and reflect on our experience. Our results suggest that student who showed positive attitude towards teamwork, benefited the most from the course and increase their perspectives on societal issues as measured by their moral reasoning after the course.
Keywords: engineering education ; soft skills ; global issues ; interdisciplinary approach ; teamwork ; communication ; climate ; food ; energy ; health ; mobility

The author manuscript is available on ResearchGate.

Posted by Antonin Danalet at 16:05
Comments (1)
Des piétons suivis à la trace grâce au WiFi - Revue de presse

Suite à la publication lundi 8 mars d'une news sur la page d'accueil de l'EPFL (et aussi dans le journal interne de l'EPFL, le Flash), quelques reprises de l'info dans les médias :

Aussi en anglais :

En allemand :

En italien :

Et en chinois :

Posted by Antonin Danalet at 18:56
Comments (1)
WiFi breadcrumbs reveal pedestrian patterns

How do students and employees choose their restaurant on campus? My thesis is presented on EPFL homepage: WiFi breadcrumbs reveal pedestrian patterns (link to the text).

The news also includes a video:

Posted by Antonin Danalet at 15:35
Comments (0)
PhD thesis online: Activity choice modeling for pedestrian facilities

My PhD thesis is now online. Its Digital Object Identifier (DOI) is doi:10.5075/epfl-thesis-6806. My supervisor was Prof. Michel Bierlaire and the Jury was composed of Prof. François Golay (president), Prof. Kay W. Axhausen, Prof. Yoram Shiftan and Prof. Francisco C. Pereira.

Activity choice modeling for pedestrian facilities
This thesis develops models of activity and destination choices in pedestrian facilities from WiFi traces. We adapt the activity-based travel demand analysis of urban mobility to pedestrians and to digital footprints. We are interested in understanding the sequence of activities and destinations of a pedestrian using discrete choice models and localization data from communication antennas. Activity and destination choice models are needed by pedestrian facilities for decision aid when building new infrastructure, modifying existing infrastructures, or locating points of interest. Understanding demand for activities is particularly important when facing an increasing number of visitors or when developing new activities, such as shopping or catering. Data from existing sensors, such as WiFi access points, are cheap and cover entire facilities, but are imprecise and lack semantics to describe moving, stopping, destinations or activities carried out at destinations. Thus, understanding pedestrian behavior first requires to observe the actual behavior and detect stops at destinations, and second to model the behavior. Part I of this thesis focuses on activity-episode sequence detection. We develop a Bayesian approach to merge raw localization data with other data sources in order to take into account the imprecision and describe activity-episode sequences. This approach generates several activity-episode sequences for a single individual. Each activity-episode sequence is associated with a probability of being the true sequence. The prior represents the attractivity of the different points of interest surrounding the measurement and allows the use of a priori information from other sources of data (register data, point-of-sale data, counting sensors, etc.). Part II proposes models for activity and destination choices. The joint choice of activity type and activity timing is modeled by seeing a sequence of activity episodes as a path in an activity network. Time is considered as discrete. Unlike traditional models, our model is not tour-based, starting and ending at the home location, since the daily ``home''activity is meaningless in our context. The choice set contains all combinations of activity types and time intervals. The number of different paths is thus very large (increasing with time resolution and disaggregation of types of activities). Inspired by route choice models, we use a Metropolis-Hastings algorithm for the sampling of paths to generate the choice set. An importance sampling correction of the utility allows the estimation of unbiased model parameters without enumerating the full choice set. While the activity path model describes the choice of an activity type in time, the location where the activity is performed is modeled with a destination choice model conditional on the activity type. Our approach accounts for the panel nature of the data and deals with serial correlation between error terms. Using real WiFi data collected on the EPFL campus, we detect pedestrian activity-episode sequences, estimate an activity path choice model and develop a destination choice model for a specific activity type: eating. Knowing that the individual has decided to eat, which restaurant does she choose? This conditional destination choice model includes in its utility the cost of menus, available types of foods and drinks, distance from a previous activity episode, socioeconomic characteristics and habits.
Keywords: Activity choice ; Destination choice ; Network traces ; Pedestrians ; Activity-episode sequence ; Importance sampling ; Dynamic model ; Initial conditions problem ; Panel data ; Location choice
Posted by Antonin D. at 14:52
Comments (0)
Ma thèse en slides

J'ai défendu publiquement ma thèse vendredi passé et je viens de publier ma présentation sur le site du labo TRANSP-OR :

Modélisation du choix d'activité pour les infrastructures piétonnes

Le but de la thèse est de mieux comprendre la demande dans les infrastructures piétonnes. On souhaite savoir où les piétons effectuent des activités, à quelle heure, pour combien de temps, dans quel ordre, et pourquoi. Pour cela, on utilise les données des antennes de communication. Peut-on répondre à ces questions à partir des données des antennes WiFi ?

Présentation en PDF

Posted by Antonin Danalet at 11:48
Comments (0)
Page : « Previous 1 2 3 4 5 6 7 8 9 10 11 Next »
RSS