Nicolas Travers is Professor of Computer Science at ESILV Engineering School at Paris la Défense. Nicolas is deputy director of the DVRC laboratory. He was habilitated (HDR) in Computer Science from Sorbonne University in 2018. His main topics deal with database management and optimization. He is specialized in distributed databases, graph databases, Information Retrieval and recommendation systems. He particularly specialized in ad hoc databases design in a transdisciplinary approach: Digital Tourism & Geography, Knowledge Management & History, IR & Musicology or Multimodal libraries.
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
Leveraging recommendations using a multiplex graph database Journal Article
In: International Journal of Web Information Systems, vol. 20, no. 5, pp. 537-582, 2024.
@article{foutse_3196,
title = {Leveraging recommendations using a multiplex graph database},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://www.emerald.com/insight/content/doi/10.1108/IJWIS-05-2024-0137/full/html},
year = {2024},
date = {2024-10-01},
journal = {International Journal of Web Information Systems},
volume = {20},
number = {5},
pages = {537-582},
abstract = {Purpose
By applying targeted graph algorithms, the method used by the authors enables effective prediction of user interactions and thus fulfils the complex requirements of modern recommender systems. This study sets a new benchmark for multidimensional recommendation strategies and offers a path towards more advanced and user-centric models.
Design/methodology/approach
To improve multidimensional data recommendation systems, multiplex graph structures are useful to capture various types of user interactions. This paper presents a novel framework that uses a graph database to compute and manipulate multiplex graphs. The approach enables flexible dimension management and increases expressive power through a specialised algebra designed for multiplex graph manipulation.
Findings
The authors compare the multiplex graph approach with traditional matrix methods, in particular random walk with restart, and show that the method not only provides deeper insights into user preferences by integrating scores from different layers of the multiplex graph, but also outperforming matrix-based approaches in most configurations. The results highlight the potential of multiplex graphs for developing sophisticated and customised recommender systems that significantly improve both performance and explainability.
Originality/value
The study provides a formal specification of a multiplex graph construction based on interaction and content-based information; and the study also developed an algebra dedicated to multiplex graphs, enabling robust and precise graph manipulations necessary for effective recommendation queries. The authors implement these algebraic operations within the Neo4j graph database system with a thorough analysis and experimentation with three different data sets, benchmarked against traditional matrix-based methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Oussama Ayoub; Christophe Rodrigues; Nicolas Travers
LoGE: an unsupervised local-global document extension generation in information retrieval for long documents Journal Article
In: International Journal of Web Information Systems, vol. 19, no. 5/6, pp. 244-262, 2023.
@article{ayoub_2411,
title = {LoGE: an unsupervised local-global document extension generation in information retrieval for long documents},
author = {Oussama Ayoub and Christophe Rodrigues and Nicolas Travers},
url = {https://www.emerald.com/insight/content/doi/10.1108/IJWIS-07-2023-0109/full/html},
year = {2023},
date = {2023-11-01},
journal = {International Journal of Web Information Systems},
volume = {19},
number = {5/6},
pages = {244-262},
abstract = {Purpose
This paper aims to manage the word gap in information retrieval (IR) especially for long documents belonging to specific domains. In fact, with the continuous growth of text data that modern IR systems have to manage, existing solutions are needed to efficiently find the best set of documents for a given request. The words used to describe a query can differ from those used in related documents. Despite meaning closeness, nonoverlapping words are challenging for IR systems. This word gap becomes significant for long documents from specific domains.
Design/methodology/approach
To generate new words for a document, a deep learning (DL) masked language model is used to infer related words. Used DL models are pretrained on massive text data and carry common or specific domain knowledge to propose a better document representation.
Findings
The authors evaluate the approach of this study on specific IR domains with long documents to show the genericity of the proposed model and achieve encouraging results.
Originality/value
In this paper, to the best of the authors' knowledge, an original unsupervised and modular IR system based on recent DL methods is introduced.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Tiange Zhu; Raphaël Fournier-S'niehotta; Philippe Rigaux; Nicolas Travers
A Framework for Content-Based Search in Large Music Collections Journal Article
In: Big Data and Cognitive Computing, vol. 6, no. 1, pp. 23, 2022.
@article{zhu_1776,
title = {A Framework for Content-Based Search in Large Music Collections},
author = {Tiange Zhu and Raphaël Fournier-S'niehotta and Philippe Rigaux and Nicolas Travers},
url = {https://www.mdpi.com/2504-2289/6/1/23},
year = {2022},
date = {2022-02-01},
journal = {Big Data and Cognitive Computing},
volume = {6},
number = {1},
pages = {23},
abstract = {We address the problem of scalable content-based search in large collections of music documents. Music content is highly complex and versatile and presents multiple facets that can be considered independently or in combination. Moreover, music documents can be digitally encoded in many ways. We propose a general framework for building a scalable search engine, based on (i) a music description language that represents music content independently from a specific encoding, (ii) an extendible list of feature-extraction functions, and (iii) indexing, searching, and ranking procedures designed to be integrated into the standard architecture of a text-oriented search engine. As a proof of concept, we also detail an actual implementation of the framework for searching in large collections of XML-encoded music scores, based on the popular ElasticSearch system. It is released as open-source in GitHub, and available as a ready-to-use Docker image for communities that manage large collections of digitized music documents},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Quentin Grossetti; Cédric Du Mouza; Nicolas Travers; Camélia Constantin
Reducing the filter bubble effect on Twitter by considering communities for recommendations Journal Article
In: International Journal of Web Information Systems, vol. 17, no. 6, pp. 728?752, 2021.
@article{grossetti_1663,
title = {Reducing the filter bubble effect on Twitter by considering communities for recommendations},
author = {Quentin Grossetti and Cédric Du Mouza and Nicolas Travers and Camélia Constantin},
url = {https://www.emerald.com/insight/content/doi/10.1108/IJWIS-06-2021-0065/full/html},
year = {2021},
date = {2021-12-01},
journal = {International Journal of Web Information Systems},
volume = {17},
number = {6},
pages = {728?752},
abstract = {Purpose
Social network platforms are considered today as a major communication mean. Their success leads to an unprecedented growth of user-generated content; therefore, finding interesting content for a given user has become a major issue. Recommender systems allow these platforms to personalize individual experience and increase user engagement by filtering messages according to user interest and/or neighborhood. Recent research results show, however, that this content personalization might increase the echo chamber effect and create filter bubbles that restrain the diversity of opinions regarding the recommended content.
Design/methodology/approach
The purpose of this paper is to present a thorough study of communities on a large Twitter data set that quantifies the effect of recommender systems on users' behavior by creating filter bubbles. The authors further propose their community-aware model (CAM) that counters the impact of different recommender systems on information consumption.
Findings
The authors propose their CAM that counters the impact of different recommender systems on information consumption. The study results show that filter bubbles effects concern up to 10% of users and the proposed model based on the similarities between communities enhance recommendations.
Originality/value
The authors proposed the CAM approach, which relies on similarities between communities to re-rank lists of recommendations to weaken the filter bubble effect for these users.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Gérald Kembellec; Nicolas Travers
XML et son écosystème Journal Article
In: Techniques de l'Ingénieur - Technologies logicielles, Architectures des systèmes, vol. TIB302DUO, no. h3502, pp. 23, 2021.
@article{kembellec_1495,
title = {XML et son écosystème},
author = {Gérald Kembellec and Nicolas Travers},
url = {https://www.techniques-ingenieur.fr/base-documentaire/technologies-de-l-information-th9/management-des-systemes-d-information-42302210/xml-et-son-ecosysteme-h3502/#presentation},
year = {2021},
date = {2021-06-01},
journal = {Techniques de l'Ingénieur - Technologies logicielles, Architectures des systèmes},
volume = {TIB302DUO},
number = {h3502},
pages = {23},
abstract = {Cet article traite de la structuration de fichiers XML, de la manière de les produire, de les utiliser, de les requêter à travers divers prismes. En effet, après une courte introduction historique sur les causes industrielles et intellectuelles qui ont amené à l'avènement d'XML comme format de stockage de données et d'informations, l'article revient sur les grammaires et vocabulaires qui permettent la structuration et la qualification documentaire dans l'industrie ou la culture. L'article se poursuit par l'application des règles du XML dans la gestion des connaissances et par une incursion dans le Web des données liées. Enfin, l'article présente XML comme structure, vecteur de stockage et de partage de données : il explore le potentiel d'XML comme base de données, les méthodes de requêtage, d'échange et de flux de données.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
David Dupuis; Cédric Du Mouza; Nicolas Travers; Gaël Chareyron
Real-Time Influence Maximization in a RTB Setting Journal Article
In: Data Science and Engineering, vol. 5, no. 3, pp. 224-239, 2020.
@article{dupuis_1233,
title = {Real-Time Influence Maximization in a RTB Setting},
author = {David Dupuis and Cédric Du Mouza and Nicolas Travers and Gaël Chareyron},
url = {https://link.springer.com/article/10.1007%2Fs41019-020-00132-2},
year = {2020},
date = {2020-09-01},
journal = {Data Science and Engineering},
volume = {5},
number = {3},
pages = {224-239},
abstract = {To maximize the impact of an advertisement campaign on social networks, the real-time bidding (RTB) systems aim at targeting the most influential users of this network. Influence maximization (IM) is a solution that addresses this issue by maximizing the coverage of the network with top-k influencers who maximize the diffusion of information. Associated with online advertising strategies at Web scale, RTB is faced with complex ad placement decisions in real time to deal with a high-speed stream of online users. To tackle this issue, IM strategies should be modified in order to integrate RTB constraints. While most traditional IM methods deal with static sets of top influencers, they hardly address the dynamic influence targeting issue by integrating short time decision, no interchange and stream's incompleteness. This paper proposes a real-time influence maximization approach which takes influence maximization decisions within a real-time bidding environment. A deep analysis of influence scores of users over several social networks is presented as well a strategy to guarantee the impact of an IM strategy in order to define the budget of an ad campaign. Finally, we offer a thorough experimental process to compare static versus dynamic IM solutions wrt. influence scores.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Hicheme Chaalal; Nicolas Travers; Hafida Belbachir
T-plotter: A new data structure to reconcile OLAP and OLTP models Journal Article
In: Multiagent and Grid Systems, vol. 15, no. 3, pp. 237-257, 2019.
@article{chaalal_885,
title = {T-plotter: A new data structure to reconcile OLAP and OLTP models},
author = {Hicheme Chaalal and Nicolas Travers and Hafida Belbachir},
url = {https://content.iospress.com/articles/multiagent-and-grid-systems/mgs190311},
year = {2019},
date = {2019-10-25},
journal = {Multiagent and Grid Systems},
volume = {15},
number = {3},
pages = {237-257},
abstract = {Classical databases represent the traditional RDBMS's and the most widely used RDBMS in the world of databases and information systems; they have been regarded as the best systems for managing data. Today with the growth of the applications and data consumers, and its openness to the general public, tradi- tional Databases are not able to meet the needs of a large number of applications, including OLAP data processing and Business Intelligence analysis; As a result, many variants of DBMS have emerged like: Column Store, In Memory and NOSQL Databases, that meet users' expectations well, and which are better adapted to cur- rent needs. As a result, the scope of classical databases has become increasingly restricted to handle OLTP models and other few models. To deal with this prob- lem, vertical fragmentation is the best way to effectively handle the OLAP model, but this technique fails to handle some analytical queries with low selectivity, pre- senting poor results in some cases. In this perspective, we propose a new vertical fragmentation design T-Plotter which makes it possible to deal effectively with the whole of analytical queries and improve the performance of RDBMSs to process the OLAP data models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Victor David; Raphaël Fournier-S'niehotta; Nicolas Travers
MAP Inference Reasoning on TMLN with Neo4j Conference
40ème conférence sur la gestion des données, Orléans, France, 2024.
@conference{david_3156,
title = {MAP Inference Reasoning on TMLN with Neo4j},
author = {Victor David and Raphaël Fournier-S'niehotta and Nicolas Travers},
url = {https://bda2024.sciencesconf.org/resource/page/id/7},
year = {2024},
date = {2024-10-01},
booktitle = {40ème conférence sur la gestion des données},
address = {Orléans, France},
abstract = {Reasoning on inconsistent and uncertain data is challenging, especially for Knowledge-Graphs (KG) to abide temporal consistency. Our goal is to enhance inference with more general time interval semantics that specify their validity, as regularly found in historical sciences. We propose a new Temporal Markov Logic Networks (TMLN) model which extends the Markov Logic Networks (MLN) modelwithuncertaintemporalfactsandrules.Totalandpartialtemporal (in)consistency relations between sets of temporal formulae are examined. We then propose a new Temporal Parametric Semantics (TPS) which allows combining several sub-functions leading to different assessment strategies. Finally, we present the NeoMaPy tool, to compute the MAP inference on MLNs and TMLNs with several TPS. We compare our performances with state-of-the-art inference tools and exhibit faster and higher quality results.},
note = {21 au 24/10/2024},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
A Multiplex Graph for Recommending Multidimensional Documents Conference
40ème conférence sur la gestion des données, Orléans, France, 2024.
@conference{foutse_3157,
title = {A Multiplex Graph for Recommending Multidimensional Documents},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://bda2024.sciencesconf.org/resource/page/id/7},
year = {2024},
date = {2024-10-01},
booktitle = {40ème conférence sur la gestion des données},
address = {Orléans, France},
abstract = {Network Science has become a flourishing interest in the last
decades with the Big Data explosion. To improve multidimensional data
Recommendation Systems, multiplex graph structures are useful to cap-
ture various types of user interactions. We propose a graph database
approach to compute multiplex graphs which helps both manipulating
dimensions in a flexible way and enhancing expressiveness with algebra
to express manipulations on the multiplex graph. Applied operations
rely on graph algorithms to predict user interactions. We compare our
approach with the traditional matrix approach with Random Walk with
Restart. The study shows that combination of scores from layers of mul-
tiplex graphs provide important insights into user preferences, with most
configurations outperforming traditional matrix methods. This approach
provides a comprehensive analysis of multidimensional recommendation
strategies in multiplex graphs, which provides capabilities of managing
different dimensions for queries, paving the way for more sophisticated
and customized recommendation systems and its explicability.},
note = {du 21 au 24/10/2024},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
How to recommend Multidimensional Documents with a Multiplex Graph? Conference
Deep Learning Indaba, Dakar, Sénégal, 2024.
@conference{foutse_3140,
title = {How to recommend Multidimensional Documents with a Multiplex Graph?},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://deeplearningindaba.com/2024/posters/},
year = {2024},
date = {2024-09-01},
booktitle = {Deep Learning Indaba},
address = {Dakar, Sénégal},
note = {du 01/09/2024 au 07/09/2024},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Hugo Alatrista Salas; Gaël Chareyron; Sonia Djebali; Imen Ouled Dlala; Nicolas Travers
SuPreME: Sequential Pattern Mining for Understanding Multiscale Seasonal Tourist Behavior Conference
GdR MADICS, Blois, France, 2024.
@conference{alatrista_salas_3045,
title = {SuPreME: Sequential Pattern Mining for Understanding Multiscale Seasonal Tourist Behavior},
author = {Hugo Alatrista Salas and Gaël Chareyron and Sonia Djebali and Imen Ouled Dlala and Nicolas Travers},
editor = {CNRS},
url = {https://www.madics.fr/event/symposium-madics-6/#ListePosters},
year = {2024},
date = {2024-05-01},
booktitle = {GdR MADICS},
address = {Blois, France},
note = {29-30/05/2024},
keywords = {},
pubstate = {online},
tppubtype = {conference}
}
Wissal Benjira; Faten Atigui; Benedicte Bucher; Malika Grim-Yefsah; Nicolas Travers
39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications, Montpellier, France, 2023.
@conference{benjira_2440,
title = {Un graphe de données pour mesurer les Objectifs de Développement Durable et comprendre l'héritage des événements sportifs sur les villes},
author = {Wissal Benjira and Faten Atigui and Benedicte Bucher and Malika Grim-Yefsah and Nicolas Travers},
url = {https://bda2023.sciencesconf.org/resource/page/id/11},
year = {2023},
date = {2023-10-01},
booktitle = {39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications},
address = {Montpellier, France},
abstract = {La mise à disposition du big data et des technologies d'observation jouent un rôle majeur dans le suivi du développement durable des villes. Produire des indicateurs mesurant les Objectifs de Développe- ment Durable (ODD) est un enjeu sociétal, nécessitant des sources hétérogènes, pour des comparaisons spatio-temporelles. Notre tra- vail se focalise particulièrement sur l'étude de l'impact des événe- ments sportifs sur les indicateurs de Développement Durable en mettant en évidence les enjeux managériaux liés à l'organisation de tels événements. Nous proposons une approche reposant sur un framework simplifié pour évaluer l'héritage laissé par ces événe- ments, en utilisant un lac sémantique représenté par une base de données graphe stockée dans Neo4j. Cette approche est illustrée par une étude de cas basée sur des données variées, permettant de comparer les indicateurs avant, pendant et après la Coupe d'Europe de Football 2016 et d'évaluer leur évolution dans différentes zones géographiques.},
keywords = {},
pubstate = {online},
tppubtype = {conference}
}
Victor David; Raphaël Fournier-S'niehotta; Nicolas Travers
NeoMaPy: calcul de MAP inference sur des graphs de connaissance temporels Conference
39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications, Montpellier, France, 2023.
@conference{david_2442,
title = {NeoMaPy: calcul de MAP inference sur des graphs de connaissance temporels},
author = {Victor David and Raphaël Fournier-S'niehotta and Nicolas Travers},
url = {https://bda2023.sciencesconf.org/resource/page/id/11},
year = {2023},
date = {2023-10-01},
booktitle = {39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications},
address = {Montpellier, France},
abstract = {Markov Logic Networks (MLN) are used for rea- soning on uncertain and inconsistent temporal data. We proposed the TMLN (Temporal Markov Logic Network) which extends them with sorts/types, weights on rules and facts, and various temporal consistencies. The NeoMaPy framework integrates it in a knowledge graph based on conflict graphs, which offers flexibility for reasoning with parame- tric Maximum A Posteriori (MAP) inferences, effi- ciency thanks to an optimistic heuristic and interac- tive graph visualization for results explanation.},
keywords = {},
pubstate = {online},
tppubtype = {conference}
}
Jihane Mali; Faten Atigui; Ahmed Azough; Nicolas Travers; Ahvar Shohreh
A Multidimensional Cost Model for Distributed Denormalized NoSQL Schemas Conference
39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications, montpellier, France, 2023.
@conference{mali_2441,
title = {A Multidimensional Cost Model for Distributed Denormalized NoSQL Schemas},
author = {Jihane Mali and Faten Atigui and Ahmed Azough and Nicolas Travers and Ahvar Shohreh},
url = {https://bda2023.sciencesconf.org/resource/page/id/11},
year = {2023},
date = {2023-10-01},
booktitle = {39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications},
address = {montpellier, France},
abstract = {The complexity of database systems has increased significantly along with the continuous growth of data, forcing Information Systems (IS) administrators to constantly adapt their data models and carefully choose the best option(s) for storing and managing data. In this context, we propose an automatic global approach for leading data model's transformation process. This approach starts with the generation of all possible solutions. It then relies on a cost model that helps to compare these generated data models to finally choose the best one for the given use case. This cost model integrates both data model and queries cost. It also takes into consideration the environmental impact of a data model as well as its financial and its time cost. This work presents for the first time a multidimensional cost model encompassing time, environmental and financial constraints, which compares data models leading to the choice of the best one for a given use case and context. In addition, a simulation tool for data model's transformation and cost computation has been developed based on our approach.},
keywords = {},
pubstate = {online},
tppubtype = {conference}
}
Pierre Lefebvre; Ahmed Azough; Nicolas Travers; Driss Yakoubi
GdR Madics, Troyes, France, 2023.
@conference{lefebvre_2306,
title = {Vid2Graph : un framework pour l'extraction de connaissances et l'analyse sémantique des flux de vidéosurveillance de systèmes distribués},
author = {Pierre Lefebvre and Ahmed Azough and Nicolas Travers and Driss Yakoubi},
url = {https://www.dataia.eu/evenements/5eme-symposium-gdr-madics},
year = {2023},
date = {2023-05-01},
booktitle = {GdR Madics},
address = {Troyes, France},
abstract = {La vidéosurveillance s'est considérablement développée ces dernières années. Les sources sont de plus en plus nombreuses, en mouvement ou avec des qualités variables : on parle de système distribué. L'analyse des données produites par de tels systèmes est devenue un enjeu majeur. En effet, si la détection des objets et des actions capturés par une caméra individuelle est aujourd'hui accessible à travers les modèles d'apprentissage automatique ou de reconnaissance de forme, la modélisation et la détection automatique d'évènements longue durée et faisant intervenir un réseau de caméras de surveillance restent un défi.
Ainsi, comment permettre la détection d'évènements complexes dans un réseau de caméras de surveillance hétérogènes et distribuées ?
Afin de répondre à cette problématique, nous proposons un framework pour l'extraction et l'enrichissement de caractéristiques à partir de caméras de vidéosurveillance. Il repose sur 1) un pipeline de modèles de Deep Learning pour l'extraction de caractéristiques de vidéos (extraction d'images-clés, détections d'objets / segmentation d'instances, extraction d'attributs, détection de relations spatiales, réidentification), 2) un module de génération d'un graphe de connaissances, 3) un module d'enrichissement du graphe pour améliorer la qualité des détections, et 4) un module d'analyse pour la détection d'événements complexes sur le graph. Son architecture modulaire permet d'interchanger les étapes d'extraction de caractéristiques provenant des vidéos. Le poster détaillera le framework proposé et illustrera le processus de création du graphe à partir de vidéos provenant du benchmark Smart-City CCTV Violence Detection Dataset (SCVD).
L'intérêt de l'approche est de pouvoir, à terme, se focaliser sur la sémantique des vidéos comme l'isolation de segments vidéo ou d'actions (filtres/projections sur le graph), la détection d'événements ou activités au moyen d'algorithmes de Graph Mining / GNN.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Wissal Benjira; Faten Atigui; Malika Grim-Yefsah; Benedicte Bucher; Nicolas Travers
GdR Madics, Troyes, France, 2023.
@conference{benjira_2308,
title = {Data lake and metadata repositories to develop city sustainability indicators using open Big Data. Applied to sports practices},
author = {Wissal Benjira and Faten Atigui and Malika Grim-Yefsah and Benedicte Bucher and Nicolas Travers},
url = {https://www.madics.fr/event/symposium-madics-5/},
year = {2023},
date = {2023-05-01},
booktitle = {GdR Madics},
address = {Troyes, France},
abstract = {La disponibilité croissante des données est une opportunité pour mieux observer et comprendre le monde réel. Produire des indicateurs mesurant les Objectifs de Developpement Durable (ODD) est un enjeux sociétal majeur, nécessitant des sources hétérogènes, pour comparer des situations dans l'espace et dans le temps. Par exemple, l'ODD11.7 s'intéresse à l'accès pour tous aux espaces publics sûrs, tels que les espaces verts, les espaces pour les pratiques sportives. Notre travail se focalise particulièrement sur l'étude de l'impact des méga-événements sportifs (ex. JO2024) sur la pratique sportive dans la ville durable et l'impact de l'organisation de méga-événements sur ces villes. L'organisation de ces événements sportifs revêt donc de multiples enjeux managériaux pour les territoires d'accueil. Selon Chappelet, les méga-événements ont le pouvoir [...] d'impacter suffisamment le territoire pour que ce dernier puisse en bénéficier durablement. Ils laissent un "héritage" sur leur passage que Preuss définit comme étant « tous les résultats qui affectent les personnes et/ou l'espace » sur la durée.
L'étude de l'impact des évènements sportifs nécessite l'exploitation de données massives connues par leur volume, variété et vélocité. Il est également indispensable de pouvoir les croiser. Seulement, il apparaît que le potentiel de ces données n'est pas entièrement exploité dû à l'absence de cadre unificateur. De ce fait, notre sujet a pour objectif de comprendre comment les données peuvent servir à mesurer l'héritage d'un méga-événement. Pour obtenir une évaluation de cet héritage, nous proposons une approche reposant sur un Framework simplifié qui facilite l'identification, le traitement ainsi que l'interprétation du croisement des données. Le lac sémantique obtenu est représenté et structuré sous forme de base de données graph facilitant le rapprochement des données avec 1) un schéma de graph de données unifié, 2) des règles d'enrichissement du graph, 3) des opérations sur le graph pour produire des mesures et leur composition pour définir les indicateurs.
Nous présenterons notre approche en nous reposant sur deux cas d'étude plus précis :
- La comparaison des parcours sportifs en ville, à vélo ou à pied, avant et après un mégaévènement à l'aide d'open data
- La comparaison des parcours sportifs en ville, à vélo ou à pied, entre deux villes en se fondant sur des données produites par différentes administrations et participants et pourvues de biais différents.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Nicolas Travers; Victor David; Raphaël Fournier-S'niehotta
Where is the Truth? From Factoids to Maximum A-Posteriori Inference on Knowledge Graphs Conference
eHealth & ethics, Pôle Léonard de Vinci Paris La Défense, France, 2023.
@conference{travers_2309,
title = {Where is the Truth? From Factoids to Maximum A-Posteriori Inference on Knowledge Graphs},
author = {Nicolas Travers and Victor David and Raphaël Fournier-S'niehotta},
url = {https://conferences.dvrc.fr/eHealth-ethics23/schedule.php},
year = {2023},
date = {2023-04-01},
booktitle = {eHealth & ethics},
address = {Paris La Défense, France},
organization = {Pôle Léonard de Vinci},
abstract = {Reasoning on uncertain and complex data to infer the most probable outcome is a crucial task in finding the
truth in complex sets of contradictory factoids (potential facts), such as confronting a hypothesis of disease
over a set of symptoms for a population. Excelling at this task involves examining logical links between
factoids, surfacing the conflicts among facts, and filtering them out to reach a consistent state called goal
recognition.
Knowledge graphs (KGs) are a key tool for modelling actions, represented as facts with a (subject,
predicate, object) triplet. Several approaches have been developed to reason on KGs, and one of them,
Markov Logic Networks [Chekol et al., 2016, Rinc ?e et al., 2018], propose probabilistic reasoning in a unified
formalism supporting uncertainty, which is key for goal recognition. However, only few MLN-based KG
reasoning frameworks handle both uncertainty and temporal data [Chekol et al., 2017], and none can handle
a fully uncertain universe where any fact or rule may be uncertain.
For goal recognition, reasoning under uncertainty and inconsistency is at the basis of methodology: the
validity of the knowledge of any hypothesis remains questionable. Temporal information is also crucial:
outside of a temporal interval, a fact becomes false. By using time intervals in knowledge, it is possible
to refine the reasoning on medical hypothesis during a sequence of events. Once all hypotheses has been
modelled, the goal recognition is to predict the most likely consistent hypothesis among all factoids/symptoms
from a population. In [Ha et al., 2011, Lester et al., 2013, Baikadi et al., 2014, Ha et al., 2014] the authors
propose MLN-based learning approaches and in [Min et al., 2014, Min et al., 2016] deep Learning approaches
to enhance goal recognition. However, those representations hardly handle a fully uncertain universe (facts
and rules). Moreover, they are not explainable, one cannot trace back the facts that leads to a given
conclusion (goal).},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Oussama Ayoub; Ludovic Li; Christophe Rodrigues; Nicolas Travers
LoGE: Expansion Locale-Globale de document non supervise avec un moteur de recherche Extensible Conference
TextMine - Groupe de travail sur la fouille de textes @ confrence EGC, Lyon, France, 2023.
@conference{ayoub_2078,
title = {LoGE: Expansion Locale-Globale de document non supervise avec un moteur de recherche Extensible},
author = {Oussama Ayoub and Ludovic Li and Christophe Rodrigues and Nicolas Travers},
url = {https://textmine.sciencesconf.org/resource/page/id/4},
year = {2023},
date = {2023-01-01},
booktitle = {TextMine - Groupe de travail sur la fouille de textes @ confrence EGC},
address = {Lyon, France},
abstract = {Avec la croissance continue des donnes textuelles que les systmes d'information modernes doivent grer, des solutions de recherche d'information sont ncessaires pour trouver efficacement le meilleur ensemble de documents pour une demande donne.Pour rsoudre ce problme, nous proposons un moteur de recherche extensible qui vise gnrer une expansion des documents en s'appuyant sur des mthodes rcentes d'apprentissage profond et mis en uvre sur Elasticsearch. Pour gnrer de nouveaux mots pour un document, un modle de langage masqu d'apprentissage profond est utilis pour infrer des mots apparents.La dmonstration montrera la fois l'extensibilit de notre cadre de gnration d'expansions, l'efficacit de l'valuation et l'impact de diverses expansions sur la correspondance des requtes.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
A recommendation system for technology intelligence based on multiplex networks Conference
17th Women in Machine Learning Workshop @ NeurIPS, New Orleans, USA, 2022.
@conference{foutse_2060,
title = {A recommendation system for technology intelligence based on multiplex networks},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://sites.google.com/view/wiml2022},
year = {2022},
date = {2022-11-01},
booktitle = {17th Women in Machine Learning Workshop @ NeurIPS},
address = {New Orleans, USA},
abstract = {A recommendation system for technology intelligence based on multiplex networks},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Oussama Ayoub; Christophe Rodrigues; Nicolas Travers
Un générateur d'extension de documents non supervisé pour moteurs de recherche Conference
GdR Traitement Automatique de la Langue, CNRS@IRISA Rennes, France, 2022.
@conference{ayoub_2079,
title = {Un générateur d'extension de documents non supervisé pour moteurs de recherche},
author = {Oussama Ayoub and Christophe Rodrigues and Nicolas Travers},
url = {https://gdr-tal-rennes.sciencesconf.org/resource/page/id/2},
year = {2022},
date = {2022-10-01},
booktitle = {GdR Traitement Automatique de la Langue},
address = {Rennes, France},
organization = {CNRS@IRISA},
abstract = {Fournir un moteur de recherche pertinent dans un contexte flexible est une tache complexe du `a lh et erog e-n eit e des donn ees (vocabulaire, taille des documents, qualit e des donn ees), leurs volumes avec des corpuscons equents. Cela pose des probl`emes sur le traitement de texte et de recherche dinformation, aussi biensur lanalyse du contenu que sur la pertinence des r esultats de requetes.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
A Technology Intelligence Recommendation System based on Multiplex Networks Conference
MaDICS 2022, CNRS Lyon, France, 2022.
@conference{foutse_1833,
title = {A Technology Intelligence Recommendation System based on Multiplex Networks},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://www.madics.fr/event/symposium-madics-4/},
year = {2022},
date = {2022-07-01},
booktitle = {MaDICS 2022},
pages = {2},
address = {Lyon, France},
organization = {CNRS},
abstract = {Mon travail de recherche se focalise sur les systèmes de recommandation pour la veille technologique chez COEXEL, une société fondée par Vincent Boisard, acteur majeur de ce domaine grâce à son outil de gestion électronique de l'information basée sur le web, appelée Mytwip®. Ces données ciblées correspondent à des évolutions technologiques visibles sur le Web pour lesquelles un expert du domaine souhaite rester informé de la concurrence ou des usages.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Nicolas Travers
MaDICS 2022, CNRS Lyon, France, 2022.
@conference{travers_1834,
title = {Panel : Représentation du contenu et extraction de connaissances à partir des textes : les systèmes de gestion de graphes, l'intelligence artificielle et les approches sémantiques},
author = {Nicolas Travers},
url = {https://www.madics.fr/event/symposium-madics-4/},
year = {2022},
date = {2022-07-01},
booktitle = {MaDICS 2022},
address = {Lyon, France},
organization = {CNRS},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Sonia Djebali; Nicolas Travers; Gaël Chareyron
Indicateurs de Mesure pour la Mobilité Touristique Conference
40ème Congrès INFORSID 2022, Dijon, France, 2022.
@conference{djebali_1832,
title = {Indicateurs de Mesure pour la Mobilité Touristique},
author = {Sonia Djebali and Nicolas Travers and Gaël Chareyron},
url = {https://inforsid2022.sciencesconf.org/},
year = {2022},
date = {2022-05-01},
booktitle = {40ème Congrès INFORSID 2022},
pages = {4},
address = {Dijon, France},
abstract = {Les traces numériques laissées par les utilisateurs sur les réseaux sociaux sont deve- nues un moyen populaire d'analyse du comportement des touristes dont l'analyse a un impact dans le marketing touristique pour construire des outils d'aide à la décision dans le tourisme. Nous proposons des mesures pour capturer la mobilité et la propagation touristique sur diffé- rentes zones qui s'appuie sur le réseau TripAdvisor et la base de données graphes Neo4j.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Oussama Ayoub; Christophe Rodrigues; Nicolas Travers
Adaptive Search Engine for Heterogeneous Documents Conference
36ème Conférence sur la Gestion de Données, virtual, 2020.
@conference{ayoub_1277,
title = {Adaptive Search Engine for Heterogeneous Documents},
author = {Oussama Ayoub and Christophe Rodrigues and Nicolas Travers},
url = {https://bda.lip6.fr/soumissions-acceptees/},
year = {2020},
date = {2020-10-01},
booktitle = {36ème Conférence sur la Gestion de Données},
address = {virtual},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Faten Atigui; Asma Mokrani; Nicolas Travers
DataGuide : une approche pour l'implantation de schémas NoSQL Conference
Extraction et Gestion de Connaissances, Bruxelles, Belgique, 2020.
@conference{atigui_1079,
title = {DataGuide : une approche pour l'implantation de schémas NoSQL},
author = {Faten Atigui and Asma Mokrani and Nicolas Travers},
url = {https://egc2020.sciencesconf.org/resource/page/id/28},
year = {2020},
date = {2020-01-01},
booktitle = {Extraction et Gestion de Connaissances},
address = {Bruxelles, Belgique},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Gaël Chareyron; Ugo Quelhas; Nicolas Travers
Neo4Tourism - A framework for Graph Data Analysis on Tourism Conference
35ème Conférence sur la Gestion de Données, Lyon, France, 2019.
@conference{chareyron_943,
title = {Neo4Tourism - A framework for Graph Data Analysis on Tourism},
author = {Gaël Chareyron and Ugo Quelhas and Nicolas Travers},
url = {https://bda.liris.cnrs.fr/list.html},
year = {2019},
date = {2019-10-01},
booktitle = {35ème Conférence sur la Gestion de Données},
address = {Lyon, France},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Nicolas Travers
Big Data Modelization: From Conception to Optimization Conference
Russian-French seminar on Big Data Applications, Paris, France, 2018.
@conference{travers_612,
title = {Big Data Modelization: From Conception to Optimization},
author = {Nicolas Travers},
url = {https://bi.hse.ru/en/rfw/2018/},
year = {2018},
date = {2018-10-01},
booktitle = {Russian-French seminar on Big Data Applications},
address = {Paris, France},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Yuehgoh Foutse; Sonia Djebali; Nicolas Travers
How to Recommend Multidimensional Data with a Multiplex Graph? Proceedings Article
In: Asian Conference on Intelligent Information and Database Systems, pp. 332-344, Springer, Ras Al Khaimah, UAE, 2024, ISBN: 978-981-97-4984-3.
@inproceedings{foutse_2898,
title = {How to Recommend Multidimensional Data with a Multiplex Graph?},
author = {Yuehgoh Foutse and Sonia Djebali and Nicolas Travers},
url = {https://aciids.pwr.edu.pl/2024/index.php#about},
issn = {978-981-97-4984-3},
year = {2024},
date = {2024-07-01},
booktitle = {Asian Conference on Intelligent Information and Database Systems},
volume = {LNAI 14796},
pages = {332-344},
publisher = {Springer},
address = {Ras Al Khaimah, UAE},
abstract = {Network Science has become a flourishing interest in the last decades with the Big Data explosion. To improve multidimensional data Recommendation Systems, multiplex graph structures are useful to capture various types of user interactions. We propose a graph database approach to compute multiplex graphs which helps both manipulating dimensions in a flexible way and enhancing expressiveness with algebra to express manipulations on the multiplex graph. Applied operations rely on graph algorithms to predict user interactions. We compare our approach with the traditional matrix approach with Random Walk with Restart. The study shows that combination of scores from layers of multiplex graphs provide important insights into user preferences, with most configurations outperforming traditional matrix methods.
This approach provides a comprehensive analysis of multidimensional recommendation strategies in multiplex graphs, which provides capabilities of managing different dimensions for queries, paving the way for more sophisticated and customized recommendation systems and its explicability.},
note = {16th Asian Conference on Intelligent Information and Database Systems
15-18 April 2024, Ras Al Khaimah, UAE},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Jihane Mali; Ahvar Shohreh; Faten Atigui; Ahmed Azough; Nicolas Travers
FACT-DM : A Framework for Automated Cost-Based Data Model Transformation Proceedings Article
In: Extending Database Technology, pp. 822–825, OpenProceedings.org, Paestum, Italy, 2024.
@inproceedings{mali_2896,
title = {FACT-DM : A Framework for Automated Cost-Based Data Model Transformation},
author = {Jihane Mali and Ahvar Shohreh and Faten Atigui and Ahmed Azough and Nicolas Travers},
url = {https://doi.org/10.48786/edbt.2024.79},
year = {2024},
date = {2024-03-01},
booktitle = {Extending Database Technology},
pages = {822--825},
publisher = {OpenProceedings.org},
address = {Paestum, Italy},
abstract = {As data continues to grow at an unprecedented rate, the com-
plexity of database systems also increase significantly, requiring
information systems (IS) architects to constantly adapt their data
model and carefully select the optimal solution(s) for storing and
managing data that align with new queries, settings, and con-
straints. Having a tool to visually show the impact of different
changes, will help IS architects in their decision making. In this
paper, we propose a framework to demonstrate the impact of
denormalization of data models on their cost and consequently
IS architects can choose the best trade-off.},
note = {25-28/03/2024},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Pierre Lefebvre; Steven Le Moal; Ahmed Azough; Nicolas Travers
NeoSGG: A Scene Graph Generation Framework for Video-Surveillance Tasks Proceedings Article
In: Extending Database Technology, pp. 838–841, OpenProceedings.org, Paestum, Italie, 2024.
@inproceedings{lefebvre_2897,
title = {NeoSGG: A Scene Graph Generation Framework for Video-Surveillance Tasks},
author = {Pierre Lefebvre and Steven Le Moal and Ahmed Azough and Nicolas Travers},
url = {https://openproceedings.org/2024/conf/edbt/paper-244.pdf},
year = {2024},
date = {2024-03-01},
booktitle = {Extending Database Technology},
pages = {838--841},
publisher = {OpenProceedings.org},
address = {Paestum, Italie},
abstract = {Video surveillance has developed considerably in the recent years. Analyzing the data generated by such systems has become a major challenge. To address this issue, we propose a framework for the creation of rich Labeled Property Graphs from video surveillance streams. It is based on 1) a Deep Learning pipeline architecture for video data extraction, 2) a graph database module to efficiently structure and store detections, and 3) a querying module to interact with generated graphs, enhancing the automatic analysis of scenes. Its modular architecture enables the feature extraction steps from the videos to be easily maintained, modified or interchanged. Our demonstration scenario shows the process of generating scene graphs from videos of several benchmark datasets. The audience will assist to an end-to-end execution of the pipeline showing the generation process and visualize generated graphs. They will have the opportunity to formulate queries using an interface illustrating several use case scenarios involving person re-identification and abandoned objects matching with their former owners.},
note = {25-28/03/2024},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Wissal Benjira; Nicolas Travers; Malika Grim-Yefsah; Faten Atigui; Benedicte Bucher
Modélisation des données ouvertes au service des ODD : application aux événements sportifs en ville Proceedings Article
In: 24ème conférence francophone sur l'Extraction et la Gestion des Connaissances, pp. 255-262, RNTI, Dijon, France, 2024.
@inproceedings{benjira_2895,
title = {Modélisation des données ouvertes au service des ODD : application aux événements sportifs en ville},
author = {Wissal Benjira and Nicolas Travers and Malika Grim-Yefsah and Faten Atigui and Benedicte Bucher},
url = {https://iutdijon.u-bourgogne.fr/egc2024/articles-acceptes/},
year = {2024},
date = {2024-01-01},
booktitle = {24ème conférence francophone sur l'Extraction et la Gestion des Connaissances},
volume = {40},
pages = {255-262},
publisher = {RNTI},
address = {Dijon, France},
abstract = {Atteindre les Objectifs de Développement Durable (ODD) est un enjeu majeur auquel nous devons faire face à l'horizon de 2030. Pour cela, de nombreux indicateurs sont indispensable et requièrent de les comparer aussi bien dans le temps et dans l'espace afin d'évaluer ces progrès. La définition de ces indicateurs nécessite l'exploitation de données existantes notamment les données ouvertes. Toutefois, la diversité des sources de données et des formats pose des défis majeurs en termes de structuration et d'intégration. Malgré l'abondance des données ouvertes, leur exploitation reste limitée, laissant un potentiel inexploité pour orienter les politiques urbaines vers la durabilité. Cet article présente une approche de modélisation dirigée par les données pour les représenter. Nous illustrons l'application de cette méthode aux ODD, et aux événements sportifs ouvrant ainsi de nouvelles perspectives pour établir le rapprochement entre données et calcul d'indicateurs.},
note = {22-26/01/2024},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Jihane Mali; Faten Atigui; Ahmed Azough; Nicolas Travers; Ahvar Shohreh
A Multidimensional Cost Model for Distributed Denormalized NoSQL Schemas Proceedings Article
In: 39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications, montpellier, France, 2023.
@inproceedings{mali_2441,
title = {A Multidimensional Cost Model for Distributed Denormalized NoSQL Schemas},
author = {Jihane Mali and Faten Atigui and Ahmed Azough and Nicolas Travers and Ahvar Shohreh},
url = {https://bda2023.sciencesconf.org/resource/page/id/11},
year = {2023},
date = {2023-10-01},
booktitle = {39ème Conférence sur la Gestion de Données - Principes, Technologies et Applications},
address = {montpellier, France},
abstract = {The complexity of database systems has increased significantly along with the continuous growth of data, forcing Information Systems (IS) administrators to constantly adapt their data models and carefully choose the best option(s) for storing and managing data. In this context, we propose an automatic global approach for leading data model's transformation process. This approach starts with the generation of all possible solutions. It then relies on a cost model that helps to compare these generated data models to finally choose the best one for the given use case. This cost model integrates both data model and queries cost. It also takes into consideration the environmental impact of a data model as well as its financial and its time cost. This work presents for the first time a multidimensional cost model encompassing time, environmental and financial constraints, which compares data models leading to the choice of the best one for a given use case and context. In addition, a simulation tool for data model's transformation and cost computation has been developed based on our approach.},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Victor David; Raphaël Fournier-S'niehotta; Nicolas Travers
NeoMaPy: a Parametric Framework for Reasoning with MAP Inference on Temporal Markov Logic Networks Proceedings Article
In: ACM, (Ed.): ACM International Conference on Information and Knowledge Management, Birmingham, UK, 2023.
@inproceedings{david_2413,
title = {NeoMaPy: a Parametric Framework for Reasoning with MAP Inference on Temporal Markov Logic Networks},
author = {Victor David and Raphaël Fournier-S'niehotta and Nicolas Travers},
editor = {ACM},
url = {https://dl.acm.org/doi/10.1145/3583780.3614757},
year = {2023},
date = {2023-10-01},
booktitle = {ACM International Conference on Information and Knowledge Management},
address = {Birmingham, UK},
abstract = {Reasoning on inconsistent and uncertain data is challenging, especially for Knowledge-Graphs (KG) to abide temporal consistency. Our goal is to enhance inference with more general time interval semantics that specify their validity, as regularly found in historical sciences. We propose a new Temporal Markov Logic Networks (TMLN) model which extends the Markov Logic Networks (MLN) model with uncertain temporal facts and rules. Total and partial temporal (in)consistency relations between sets of temporal formulae are examined. We then propose a new Temporal Parametric Semantics (TPS) which allows combining several sub-functions leading to different assessment strategies. Finally, we present the new NeoMaPy tool, to compute the MAP inference on MLNs and TMLNs with several TPS. We compare our performances with state-of-the-art inference tools and exhibit faster and higher quality results.},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Victor David; Raphaël Fournier-S'niehotta; Nicolas Travers
NeoMaPy: a Framework for Computing MAP inference on Temporal Knowledge Graphs Proceedings Article
In: International Joint Conference on Artificial Intelligence, pp. 7123-7126, Macao, China, 2023.
@inproceedings{david_2307,
title = {NeoMaPy: a Framework for Computing MAP inference on Temporal Knowledge Graphs},
author = {Victor David and Raphaël Fournier-S'niehotta and Nicolas Travers},
url = {https://www.ijcai.org/proceedings/2023/831},
year = {2023},
date = {2023-08-01},
booktitle = {International Joint Conference on Artificial Intelligence},
pages = {7123-7126},
address = {Macao, China},
abstract = {Markov Logic Networks (MLN) are used for rea- soning on uncertain and inconsistent temporal data. We proposed the TMLN (Temporal Markov Logic Network) which extends them with sorts/types, weights on rules and facts, and various temporal consistencies. The NeoMaPy framework integrates it in a knowledge graph based on conflict graphs, which offers flexibility for reasoning with parame- tric Maximum A Posteriori (MAP) inferences, effi- ciency thanks to an optimistic heuristic and interac- tive graph visualization for results explanation.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Julien Martin-Prin; Imen Ouled Dlala; Nicolas Travers; Said Jabbour
A Distributed SAT-based Framework for Closed Frequent Itemset Mining Proceedings Article
In: International Conference on Advanced Data Mining and Applications, Springer, Brisbane, Australia, 2022.
@inproceedings{martin-prin_1890,
title = {A Distributed SAT-based Framework for Closed Frequent Itemset Mining},
author = {Julien Martin-Prin and Imen Ouled Dlala and Nicolas Travers and Said Jabbour},
url = {https://adma2022.uqcloud.net/important_date.html},
year = {2022},
date = {2022-11-01},
booktitle = {International Conference on Advanced Data Mining and Applications},
publisher = {Springer},
address = {Brisbane, Australia},
abstract = {Frequent Itemset Mining is an essential part of data mining. SAT-based approaches that extract frequent itemsets in big data encounter significant challenges when computing power and storage capacity are limited. This paper proposes an efficient distributed SAT-based framework for the Closed Frequent Itemset Mining problem (CFIM) which minimizes communications throughout the distributed architecture and reduces bottlenecks due to shared memory. Moreover, it enhances scalability and fault tolerance. This approach makes use of a Computation-Distributed Paradigm to enumerate efficiently the set of all closed itemsets, by reducing the processing time. To the best of our knowledge, this paper presents the first attempt towards a distributed SAT-based approach for CFIM. An extensive empirical evaluation on various real-word datasets shows the efficiency of the approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Hugo Prevoteau; Sonia Djebali; Zhao Laiping; Nicolas Travers
Propagation Measure on Circulation Graphs for Tourism Behavior Analysis Proceedings Article
In: Bases de Données Avancées, Clermont-Ferrand, France, 2022.
@inproceedings{prevoteau_1921,
title = {Propagation Measure on Circulation Graphs for Tourism Behavior Analysis},
author = {Hugo Prevoteau and Sonia Djebali and Zhao Laiping and Nicolas Travers},
url = {https://bda2022.sciencesconf.org/resource/page/id/13},
year = {2022},
date = {2022-10-01},
booktitle = {Bases de Données Avancées},
address = {Clermont-Ferrand, France},
abstract = {Social network analysis has widespread in recent years, especially in digital tourism. Indeed the large amount of data that tourists produce during their travels represents an effective source to un- derstand their behavior and is of great importance for tourism stakeholders. This paper studies the propagation effect of tourists on the territory thanks to geotagged circulation graphs.
A new weighted measure is introduced for circulation charac- terization based on both topologies and distances. This measure helps to determine the behavior of tourists on local and global areas. An optimization strategy based on spanning trees is applied to reduce the computation on the whole graph while keeping a good approximation of the behavior.},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Jihane Mali; Ahvar Shohreh; Faten Atigui; Ahmed Azough; Nicolas Travers
A Global Model-Driven Denormalization Approach for Schema Migration Proceedings Article
In: International Conference on Research Challenges in Information Science, pp. 529-545, Springer, Barcelona, Spain, 2022.
@inproceedings{mali_1804,
title = {A Global Model-Driven Denormalization Approach for Schema Migration},
author = {Jihane Mali and Ahvar Shohreh and Faten Atigui and Ahmed Azough and Nicolas Travers},
url = {https://link.springer.com/chapter/10.1007/978-3-031-05760-1_31},
year = {2022},
date = {2022-05-01},
booktitle = {International Conference on Research Challenges in Information Science},
volume = {446},
pages = {529-545},
publisher = {Springer},
address = {Barcelona, Spain},
abstract = {Abstract. With data's evolution in terms of volume, variety, and ve- locity, Information Systems (IS) administrators have to steadily adapt their data model and choose the best solution(s) to store and manage data in accordance with users' requirements. In this context, many exist- ing solutions transform a source data model into a target one, but none of them leads the administrator to choose the most suitable model by offering a limited solution space automatically calculated and adapted to his needs. We propose ModelDrivenGuide, an automatic global approach for leading the model transformation process. It starts by transforming the conceptual model into a logical model, and it defines refinement rules that help to generate all possible data models. Our approach then relies on a heuristic to reduce the search space by avoiding cycles and redun- dancies. We also propose a formalisation of the denormalization process and we discuss the completeness and the complexity of our approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Elena Chatzopoulou; Nicolas Travers
Authenticity Goes Digital: A Big Data Analysis of the Influence of the Country of Origin and Authenticity Perceptions on TripAdvisor Ethnic Restaurant Reviews Proceedings Article
In: Academy of Marketing Science Annual Conference, pp. 3-15, Springer, online, 2022.
@inproceedings{chatzopoulou_2883,
title = {Authenticity Goes Digital: A Big Data Analysis of the Influence of the Country of Origin and Authenticity Perceptions on TripAdvisor Ethnic Restaurant Reviews},
author = {Elena Chatzopoulou and Nicolas Travers},
url = {https://link.springer.com/chapter/10.1007/978-3-030-89883-0_2},
year = {2022},
date = {2022-05-01},
booktitle = {Academy of Marketing Science Annual Conference},
pages = {3-15},
publisher = {Springer},
address = {online},
abstract = {Authenticity perceptions are subjectively driven and rely on social constructions making the concept hard to be defined. The current study is following a big data approach to capture perceptions and beliefs concerning the authenticity of ethnic restaurants and also when online positive reviews are given about authenticity under the influence of a visit to the country of origin. The key idea of our method relies on the analysis of a 3-step characterization of a big data repository extracted from TripAdvisor. Step 0 concerns reviews made for Italian restaurants before consumers visit Italy, step 1 concerns the reviews made while consumers were in Italy and step 2 concerns reviews made after they visited Italy. This characterization exploits both sentiment analysis and graph data models. Our findings propose a depiction of authenticity for ethnic restaurants via e-word of mouth. With a big data analysis on TripAdvisor, we provided an analysis on both ratings and comments which showed the impact of authenticity. As such, consumers, after visiting the country of origin, were more critical while they provided lower ratings and they were also focusing more on authentic atmosphere and service, showing evolution of their online reviews.},
note = {25-27/05/2022},
keywords = {},
pubstate = {online},
tppubtype = {inproceedings}
}
Hugo Prevoteau; Sonia Djebali; Zhao Laiping; Nicolas Travers
Propagation Measure on Circulation Graphs for Tourism Behavior Analysis Proceedings Article
In: ACM Symposium on Applied Computing, ACM, Brno, Czec Republic, 2022.
@inproceedings{prevoteau_1749,
title = {Propagation Measure on Circulation Graphs for Tourism Behavior Analysis},
author = {Hugo Prevoteau and Sonia Djebali and Zhao Laiping and Nicolas Travers},
url = {https://www.sigapp.org/sac/sac2022/index.html},
year = {2022},
date = {2022-04-01},
booktitle = {ACM Symposium on Applied Computing},
publisher = {ACM},
address = {Brno, Czec Republic},
abstract = {Social network analysis has widespread in recent years, especially in digital tourism. Indeed the large amount of data that tourists produce during their travels represents an effective source to understand their behavior and is of great importance for tourism stakeholders. This paper studies the propagation effect of tourists on the territory thanks to geotagged circulation graphs. Those graphs reflect traffic flows which need to be analyzed over time and space. A new weighted measure is introduced for circulation characterization based on both topologies and distances. This measure helps to determine the behavior of tourists on local and global areas. An optimization strategy based on spanning trees is applied to reduce the computation on the whole graph while keeping a good approximation of the behavior. The approach is simulated on various graphs and evaluated experimentaly over a real dataset at various geographic zones, scales, communities, and time.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Sonia Djebali; Nicolas Loas; Nicolas Travers
Indicators for Measuring Tourist Mobility Proceedings Article
In: International Conference on Web Information Systems Engineering, Amsterdam and Leiden, Netherlands, 2020, ISBN: 978-3-030-62004-2.
@inproceedings{djebali_1273,
title = {Indicators for Measuring Tourist Mobility},
author = {Sonia Djebali and Nicolas Loas and Nicolas Travers},
url = {http://wasp.cs.vu.nl/WISE2020/accept.html},
issn = {978-3-030-62004-2},
year = {2020},
date = {2020-10-01},
booktitle = {International Conference on Web Information Systems Engineering},
address = {Amsterdam and Leiden, Netherlands},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Jihane Mali; Faten Atigui; Ahmed Azough; Nicolas Travers
How to Implement NoSQL Schemas with ModelDrivenGuide? Proceedings Article
In: 36ème conférence sur la Gestion de données, virtual, 2020.
@inproceedings{mali_1381,
title = {How to Implement NoSQL Schemas with ModelDrivenGuide?},
author = {Jihane Mali and Faten Atigui and Ahmed Azough and Nicolas Travers},
url = {https://bda.lip6.fr/programme/},
year = {2020},
date = {2020-10-01},
booktitle = {36ème conférence sur la Gestion de données},
address = {virtual},
abstract = {With the evolution of data in terms of volume, variety and velocity, designing and developing an Information Systems (IS) requires studying the best solutions to store and manipulate data while respecting the user's requirements. In this demonstration, we show how to implement an IS using the ModelDrivenGuide, which is a semi-automated approach based on transformation rules starting from a conceptual model, then going from one logical model to an other by re?nement to finally the chosen physical model.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Jihane Mali; Faten Atigui; Ahmed Azough; Nicolas Travers
ModelDrivenGuide: An Approach for Implementing NoSQL Schemas Proceedings Article
In: The 31st International Conference on Database and Expert Systems Applications DEXA 2020, Bratislava, Slovakia, 2020.
@inproceedings{mali_1274,
title = {ModelDrivenGuide: An Approach for Implementing NoSQL Schemas},
author = {Jihane Mali and Faten Atigui and Ahmed Azough and Nicolas Travers},
url = {https://www.dexa.org/previous/dexa2020/dexa2020.html},
year = {2020},
date = {2020-09-01},
booktitle = {The 31st International Conference on Database and Expert Systems Applications DEXA 2020},
address = {Bratislava, Slovakia},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Gaël Chareyron; Ugo Quelhas; Nicolas Travers
Tourism Analysis on Graphs with Neo4Tourism Proceedings Article
In: International Conference on Web Information Systems Engineering, Hong Kong, China, 2020, ISBN: 978-981-15-3280-1.
@inproceedings{chareyron_1070,
title = {Tourism Analysis on Graphs with Neo4Tourism},
author = {Gaël Chareyron and Ugo Quelhas and Nicolas Travers},
url = {https://wise2019.comp.polyu.edu.hk/
https://link.springer.com/chapter/10.1007%2F978-981-15-3281-8_4},
issn = {978-981-15-3280-1},
year = {2020},
date = {2020-02-01},
booktitle = {International Conference on Web Information Systems Engineering},
address = {Hong Kong, China},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Philippe Rigaux; Nicolas Travers
Scalable Searching and Ranking for Melodic Pattern Queries Proceedings Article
In: 20th annual conference of the International Society for Music Information Retrieval, Delft, Netherlands, 2019.
@inproceedings{rigaux_888,
title = {Scalable Searching and Ranking for Melodic Pattern Queries},
author = {Philippe Rigaux and Nicolas Travers},
url = {https://ismir2019.ewi.tudelft.nl/},
year = {2019},
date = {2019-11-01},
booktitle = {20th annual conference of the International Society for Music Information Retrieval},
address = {Delft, Netherlands},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
David Dupuis; Cédric Du Mouza; Nicolas Travers; Gaël Chareyron
RTIM: a Real-Time Influence Maximization Strategy Proceedings Article
In: International Conference on Web Information Systems Engineering, pp. pp 277-292, Hong-Kong, China, 2019, ISBN: ISBN : 978-3-030-34222-7.
@inproceedings{dupuis_942,
title = {RTIM: a Real-Time Influence Maximization Strategy},
author = {David Dupuis and Cédric Du Mouza and Nicolas Travers and Gaël Chareyron},
url = {https://link.springer.com/book/10.1007/978-3-030-34223-4},
issn = {ISBN : 978-3-030-34222-7},
year = {2019},
date = {2019-11-01},
booktitle = {International Conference on Web Information Systems Engineering},
volume = {11881},
pages = {pp 277-292},
address = {Hong-Kong, China},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Quentin Grossetti; Cédric Du Mouza; Nicolas Travers
Community-based Recommendations on Twitter: Avoiding The Filter Bubble Proceedings Article
In: Web Information Systems Engineering - WISE 2019, pp. pp 212-227, Hong-Kong, China, 2019, ISBN: ISBN : 978-3-030-34222-7.
@inproceedings{grossetti_944,
title = {Community-based Recommendations on Twitter: Avoiding The Filter Bubble},
author = {Quentin Grossetti and Cédric Du Mouza and Nicolas Travers},
url = {https://link.springer.com/chapter/10.1007/978-3-030-34223-4_14},
issn = {ISBN : 978-3-030-34222-7},
year = {2019},
date = {2019-11-01},
booktitle = {Web Information Systems Engineering - WISE 2019},
volume = {11881},
pages = {pp 212-227},
address = {Hong-Kong, China},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
David Dupuis; Cédric Du Mouza; Nicolas Travers; Gaël Chareyron
Influence Maximization in a Real-Time Bidding Environment Proceedings Article
In: 35ème Conférence sur la Gestion de Données, Lyon, France, 2019.
@inproceedings{dupuis_941,
title = {Influence Maximization in a Real-Time Bidding Environment},
author = {David Dupuis and Cédric Du Mouza and Nicolas Travers and Gaël Chareyron},
url = {https://bda.liris.cnrs.fr/list.html},
year = {2019},
date = {2019-10-01},
booktitle = {35ème Conférence sur la Gestion de Données},
address = {Lyon, France},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Quentin Grossetti; Camélia Constantin; Cédric Du Mouza; Nicolas Travers
Reducing Filter Bubbles With a Community Aware Model Proceedings Article
In: 34ème Conférence sur la Gestion de Données, Bucarest, Romania, 2018.
@inproceedings{grossetti_563,
title = {Reducing Filter Bubbles With a Community Aware Model},
author = {Quentin Grossetti and Camélia Constantin and Cédric Du Mouza and Nicolas Travers},
url = {https://bda2018.ensea.fr/},
year = {2018},
date = {2018-10-01},
booktitle = {34ème Conférence sur la Gestion de Données},
address = {Bucarest, Romania},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mihaela Axente; Didier Decoupigny; Jeanne Kostrz; Charles Minart; Gaël Chareyron; Sébastien Jacquot; Mélanie Mondo; Nicolas Travers
Les pratiques touristiques dans la destination Lille métropole : Analyse à partir des traces numériques des visiteurs Miscellaneous
MEL, 2024.
@misc{axente_2969,
title = {Les pratiques touristiques dans la destination Lille métropole : Analyse à partir des traces numériques des visiteurs},
author = {Mihaela Axente and Didier Decoupigny and Jeanne Kostrz and Charles Minart and Gaël Chareyron and Sébastien Jacquot and Mélanie Mondo and Nicolas Travers},
url = {https://diffuweb.lillemetropole.fr/obstourisme/ETUDES/La%20destination%20Lille%20M%c3%a9tropole%20vue%20par%20les%20traces%20numeriques_Mai%202024.pdf},
year = {2024},
date = {2024-06-01},
pages = {1-24},
howpublished = {MEL},
keywords = {},
pubstate = {online},
tppubtype = {misc}
}
Oussama Ayoub; Nicolas Travers
CNAM, 2023.
@phdthesis{ayoub_2785,
title = {Enrichissement sémantique non supervisé de longs documents spécialisés pour la recherche d'information},
author = {Oussama Ayoub and Nicolas Travers},
url = {https://theses.fr/s234629},
year = {2023},
date = {2023-12-01},
school = {CNAM},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
No posts by this author.
