Cécile BOTHOREL

Position

Enseignant Chercheur

Campus

Brest

Contact information:

Phone

+33 2 29 00 12 40

Office information

Mme Ghislaine Le Gall
Département LUSSI
Technopôle Brest-Iroise CS 83818
29238 Brest Cedex 3

Mail

Département LUSSI
Technopôle Brest-Iroise CS 83818
29238 Brest Cedex 3
Biography

 

Cécile Bothorel received her Engineer Diploma in Computer Science in 1995, her PhD degree in Computer Science, Distributed Artificial Intelligence (Orange Labs, Rennes; CNRS/LAAS - Université Paul Sabatier, Toulouse). During her PhD, she worked on Multi-Agent Systems and Social Recommender Systems
She then worked as a researcher at Orange Labs (Lannion) for 10 years where she conducted various research in Social Media Analysis, Machine Learning and Recommender Systems. She directed phd students and postdocs, conceived and developped prototypes in the contexts of Knowledge Management, eLearning, and more generally Knowledge Sharing. She is the author of 4 patents.

In 2009, she joined the LUSSI department in IMT Atlantique, Brest (ex Telecom Bretagne). She is an Associate Professor of Computer Science, teaching Programming, Graph Theory, Social Networks Analysis, Hadoop and Big Data Analytics. She is involved in the Master of Science in Computer Science and Decision Systems (CSDS) and also in various teaching in Creativity, Innovation and Entrepreneurship (e.g. First program,  Fondation Telecom).

She is in charge of the Data Science one-year specialization of the Master of Science in Engineering "Diplôme d'Ingénieur", and is especially in charge of Data Science courses and student projects.

As a member of UMR CNRS 6285 Lab-STICC, she works on Complex Networks Analysis and Graph Mining, focusing more particularly on the Detection of Communities over the social web and how to help practionners to understand and use clustering algorithms over Social Networks. The application areas are various, in Computer Science (algorithms and a toolbox to study graphs with attributes) and in pluridisciplinary fields, e.g. Management (study of the promotion process within Wikipedia administration), Economics (study of the impact of social interactions in the success of crowdfounded projects) or Ecology (study of the topology of networks of habitats of sea species and their robustness to global warming).

She has been an expert for various innovative start-ups and the evaluation of research proposals for the ANR (French National Agency for Research programs). She is a Programme Committee Member for several international and national conferences (EGC, JFGG, MARAMI, DS, BESC, ECML PKDD, WWW, ...) and reviews (IJSNM, SNAM, IJPRAI, ACM TIST, ...).

 

PhD students

2018- Mounir Haddad. Deep learning and video mouvement detection in interaction networks: analysis of communities dynamics. Co-directed with Philippe Lenca, IMT Atlantique and Dominique Bedart, Mikaël Campion, DSI Global Services

2015-2018 Vinh-Loc Dao. Community detection and evaluation in complex networks: a descriptive approach. Co-directed with Philippe Lenca, IMT Atlantique

2011-2014 Duc Kinh Le Tran. Social Media analysis to enhance the Customer Relationship. Co-directed with Pascal Cheung, Orange Labs, Lannion.

Dec 2009-Dec 2012 Juan David Gomez Cruz. Point of view Visualisation of socio-semantic networks. Co-directed with François Poulet, IRISA, Rennes 1.

2007-2010 Damien Poirier. Opinion Mining on a Website for Recommendation systems. Co-directed with Patrick Gallinari, LIP6, Paris 6 and Isabelle Tellier, LIFO, Université d'Orléans, and Françoise Fessant, France Telecom R&D.

2004-2009 Emilie Marquois-Ogez. Social Simulation of the participation on a mailing-list. Co-directed with Frédéric Amblard, IRIT, Toulouse.

2000-2003 Karine Chevalier. Automatic profil learning from browsing logs  on a web site. Co-directed with Vincent Corruble and Jean-Gabriel Ganascia, LIP6, équipe APA, Paris 6.

 

Post Doc students

2019-2020 Raphaël Charbey. Detection of temporal motifs in temporal social networks.

2012-2015 Romain Picot-Clemente. Social network analysis to enhance places recommendation in location-based social network. Large scale study and deployment of algorithms (MapReduce Hadoop/Mahout) with Sean Chalmers (Intership).

2008 Mohamed Bouklit. Detection of communities in hypergraphs. Dynamic tag clouds functionality regarding the current request within a Prototype of cultural goods exploration. NeTopic project, Orange Explocentre.

2006 Vincent Dubois. Complex network analysis and graph-mining. Socializing Knowledge Management prototype, Knowledge Management Project, France Telecom R&D. P2P networks visualisation for administration prototype. Freddy Project, France Telecom R&D. Graph management java-library toolkit, France Telecom R&D.

 

Graduate Students

Juil. 2013 - déc. 2013. Sean Chalmers. Investigation into the current state of the art with respect to 'Big Data' frameworks and libraries. Publication of a technical detailed report with recommendations according to the practioners' needs: volume, variety, variability, velocity, people (skills needed), infrastructure. Co-directed by Romain Picot-Clemente, postdoc fellow.

Jan 2011 - Sept. 2011 Duc Kinh Le Tran. Conception et développement de fonctionnalités innovantes sociales pour un système de recommandation d'émissions de TV. Co-encadré par Pascal Cheung, Orange Labs, Lannion.

Oct 2009 - April 2010 Owen Sacco. Exploiting web2.0 folksonomy mining for Web Social Semantic Information Retrieval problematics.

2008 Mamadou Dembélé. Automatic Opinion Profile updating from textual reviews analysis. FP6 IST Pharos Project.

2007 Julien Lejeune. BuzzAround, a buzz spreading web site. Open Orange 2.0 program.

2005-2006 Cyrille Dechambenoit. Visualisation of the dynamics of social networks through emails exchange. Knowledge Science Research Program, France Telecom R&D.

2003-2004 Marc Siramy, Anna Moalic. Visualisation and exploration of eLearners social networks. Knowledge Management Project, France Telecom R&D. Formatis Project, France Telecom R&D.

 

 

 

Projet PIL (2018-2021) Projet ANR PIL

Project funded by the French ANR.

Analyse and assess the socio-economic effects of digital transformations on information quality and pluralism (QPI) in the media universe.

http://www.anr-pil.org

 

Past Projects

 

Pay2You Places (10/2011 - 04/2014)

Projet FUI12 supported by the 'pôles Finance Innovation', 'Transactions électroniques Sécurisées' et 'Images et Réseaux'.

Pay2You Places explores new payment usages via smartphones, from the payment solution to social recommendation for final users, and from an integrated payment solution to geomarketing support for professionals.

 

VIPEER (11/2009 - 11/2012)

 

Project funded by ANR Verso on Distributed Content Delivery Networks for Intra-domain Video Delivery, labelized by competitivity poles Images et Réseaux and SCS (Solutions Communicantes et Sécurisées).

The main objective of the present project is to provide methods allowing a network operator to have explicit control on traffic flows related to video distribution. VIPEER builds upon the collaboration between a traditional CDN and a peer-assisted CDN or "distributed CDN" (dCDN), i.e. an overlay controlled by the network operator using P2P paradigms. The peers in the dCDN may be network elements such as network nodes or boxes located at customers' premises.

http://recherche.telecom-bretagne.eu/vipeer/

Mazadoo2.0 (2010-2012)

Project supported by the 'pôle de compétitivité mondial Images et Réseaux' and the 'Direction Générale de la compétitivité, de l’industrie et des services'.

One of the biggest challenges we currently face is to keep elderly people immersed in their social environment when they leave their home and enter a retirement home. Many of them feel isolated. The TV stands as their favorite media, and our previous experiments showed that listening to vocalized local news and receiving TV messages and photos from family helped in fighting these feelings of isolation. Within this project, by using online social networks, we wish to involve the elderly in new types of interactions, more various and frequent. They will be more active and included in micro-conversations around multimedia contents. The retirement homes will benefit also from social networking capabilities. They will participate to the local news dedicated to the elderly people. In addition, the remote family will be informed of activities through an agenda and various publications.

http://www.mazadoo.eu

 

 

My main research interests embrace, but are not limited to, the following topics: Complex Networks Analysis, Machine Learning and Graph Mining, focusing more particularly on graphs with attributes.

  • Detection of implicit communities (clustering) in networks with attributes where both the structure of the network and the profiles of the people are combined to detect close and similar people
  • Evolution of communities, dynamics of interactions
  • Geolocalized social interactions to enhance remmender systems
  • Datamining and network metrics to measure engagement in social media (for CRM, for eMarketing)
  • Algorithms, visualization, toolbox for graphs with attributes analysis
  • Application in Recommender Systems, Marketing, Management, Economics, etc.

 

Main courses

  • Data Science and Graph Theory applied to Social Networks Analysis through 3 challenges: How to  sign a fulltime job contract with Google (weak ties theory), How to fill missing profil data on LinkedIn (Homophily), How to stop Zombies invasion (Epidemics).
  • IA and Graph Theory in Python (PyRat): Shortest paths, Travelling Salesman Problem to win a search competition in a maze.
  • Introduction to Big Data: HDFS, Map Reduce
  • Data Science, form Tools to Valorization: How to formulate a problem, Describe data, Select relevant variables, Implement the analysis tasks with Python and/or Hadoop Spark, Evaluate the analysis and Valorize the results to a Manager.

 

Tutoring

  • Innovation/Entrepreneurship: Brainstorming, Lean Start-up, Business Model, Pitch
  • Tutoring of various student Projects (5 to 10 per year)
  • Tutoring of Internship Student (5 to 10 per year)
Publications
Pré-publication, Document de travail
Dao Vinh-Loc, Bothorel Cécile, Lenca Philippe
Community structure: A comparative evaluation of community detection methods
2019
Bibtext :
@unpublished{dao:hal-01976587,
TITLE = {{Community structure: A comparative evaluation of community detection methods}},
AUTHOR = {Dao, Vinh-Loc and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01976587},
NOTE = {working paper or preprint},
YEAR = {2019},
MONTH = Jan,
KEYWORDS = {community detection ; community structure ; comparative analysis ; empirical analysis ; computation time ; community size ; structural quality function ; validation metric ; decision-making assistance for practionners},
PDF = {https://hal.archives-ouvertes.fr/hal-01976587/file/1812.06598.pdf},
HAL_ID = {hal-01976587},
HAL_VERSION = {v1},
}
Endnote :
%0 Unpublished work
%T Community structure: A comparative evaluation of community detection methods
%+ Lab-STICC_IMTA_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%A Dao, Vinh-Loc
%A BOTHOREL, Cécile
%A Lenca, Philippe
%8 2019-01-10
%D 2019
%Z 1812.06598
%K community detection
%K community structure
%K comparative analysis
%K empirical analysis
%K computation time
%K community size
%K structural quality function
%K validation metric
%K decision-making assistance for practionners
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Social and Information Networks [cs.SI]Preprints, Working Papers, ...
%X Discovering community structure in complex networks is a mature field since a tremendous number of community detection methods have been introduced in the literature. Nevertheless, it is still very challenging for practioners to determine which method would be suitable to get insights into the structural information of the networks they study. Many recent efforts have been devoted to investigating various quality scores of the community structure, but the problem of distinguishing between different types of communities is still open. In this paper, we propose a comparative, extensive and empirical study to investigate what types of communities many state-of-the-art and well-known community detection methods are producing. Specifically, we provide comprehensive analyses on computation time, community size distribution, a comparative evaluation of methods according to their optimisation schemes as well as a comparison of their partioning strategy through validation metrics. We process our analyses on a very large corpus of hundreds of networks from five different network categories and propose ways to classify community detection methods, helping a potential user to navigate the complex landscape of community detection.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01976587/document
%2 https://hal.archives-ouvertes.fr/hal-01976587/file/1812.06598.pdf
%L hal-01976587
%U https://hal.archives-ouvertes.fr/hal-01976587
%~ CNRS
%~ UNIV-BREST
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ ENIB
%~ LAB-STICC
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA
Communication dans un congrès
Dao Vinh-Loc, Bothorel Cécile, Lenca Philippe
Estimating the similarity of community detection methods based on cluster size distribution
Complex Networks 2018, The 7th International Conference on Complex Networks and Their Applications, Dec 2018, Cambridge, United Kingdom. pp.183-194
Bibtext :
@inproceedings{dao:hal-01911077,
TITLE = {{Estimating the similarity of community detection methods based on cluster size distribution}},
AUTHOR = {Dao, Vinh-Loc and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01911077},
BOOKTITLE = {{Complex Networks 2018, The 7th International Conference on Complex Networks and Their Applications}},
ADDRESS = {Cambridge, United Kingdom},
PUBLISHER = {{Springer}},
SERIES = {Studies in Computational Intelligence},
VOLUME = {812},
PAGES = {183-194},
YEAR = {2018},
MONTH = Dec,
KEYWORDS = {comparative analysis ; community detection ; similarity metric ; community size},
PDF = {https://hal.archives-ouvertes.fr/hal-01911077/file/Cluster_size_distribution_comparisation_algos_detection_community_Dao_Bothorel_Lenca_2018.pdf},
HAL_ID = {hal-01911077},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Estimating the similarity of community detection methods based on cluster size distribution
%+ Lab-STICC_IMTA_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%A Dao, Vinh-Loc
%A BOTHOREL, Cécile
%A Lenca, Philippe
%< avec comité de lecture
%B Complex Networks 2018, The 7th International Conference on Complex Networks and Their Applications
%C Cambridge, United Kingdom
%I Springer
%3 Studies in Computational Intelligence
%V 812
%P 183-194
%8 2018-12-11
%D 2018
%K comparative analysis
%K community detection
%K similarity metric
%K community size
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]
%Z Computer Science [cs]/Discrete Mathematics [cs.DM]
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Social and Information Networks [cs.SI]Conference papers
%X Detecting community structure discloses tremendous information about complex networks and unlock promising applied perspectives. Accordingly, a numerous number of community detection methods have been proposed in the last two decades with many rewarding discoveries. Notwithstanding, it is still very challenging to determine a suitable method in order to get more insights into the mesoscopic structure of a network given an expected quality, especially on large scale networks. Many recent efforts have also been devoted to investigating various qualities of community structure associated with detection methods, but the answer to this question is still very far from being straightforward. In this paper, we propose a novel approach to estimate the similarity between community detection methods using the size density distributions of communities that they detect. We verify our solution on a very large corpus of networks consisting in more than a hundred networks of five different categories and deliver pairwise similarities of 16 state-of-the-art and well-known methods. Interestingly, our result shows that there is a very clear distinction between the partitioning strategies of different community detection methods. This distinction plays an important role in assisting network analysts to identify their rule-of-thumb solutions.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01911077/document
%2 https://hal.archives-ouvertes.fr/hal-01911077/file/Cluster_size_distribution_comparisation_algos_detection_community_Dao_Bothorel_Lenca_2018.pdf
%L hal-01911077
%U https://hal.archives-ouvertes.fr/hal-01911077
%~ CNRS
%~ UNIV-BREST
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ ENIB
%~ TDS-MACS
%~ LAB-STICC
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA
Pré-publication, Document de travail
Dao Vinh-Loc, Bothorel Cécile, Lenca Philippe
An empirical characterization of community structures in complex networks using a bivariate map of quality metrics
2018
Details
Quote
Bibtext :
@unpublished{dao:hal-01809064,
TITLE = {{An empirical characterization of community structures in complex networks using a bivariate map of quality metrics}},
AUTHOR = {Dao, Vinh-Loc and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal-imt-atlantique.archives-ouvertes.fr/hal-01809064},
NOTE = {18 pages, 12 figures, 41 reference items},
YEAR = {2018},
MONTH = Jun,
HAL_ID = {hal-01809064},
HAL_VERSION = {v1},
}
Endnote :
%0 Unpublished work
%T An empirical characterization of community structures in complex networks using a bivariate map of quality metrics
%+ Lab-STICC_IMTA_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%+ Lab-STICC_TB_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (LUSSI)
%A Dao, Vinh-Loc
%A BOTHOREL, Cécile
%A Lenca, Philippe
%Z 18 pages, 12 figures, 41 reference items
%8 2018-06-06
%D 2018
%Z 1806.01386
%Z Computer Science [cs]/Social and Information Networks [cs.SI]
%Z Physics [physics]/Physics [physics]/Physics and Society [physics.soc-ph]
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]Preprints, Working Papers, ...
%X Community detection emerges as an important task in the discovery of network mesoscopic structures. However, the concept of a "good" community is very context-dependent and it is relatively complicated to deduce community characteristics using available community detection techniques. In reality, the existence of a gap between structural goodness quality metrics and expected topological patterns creates a confusion in evaluating community structures. In this paper, we introduce an empirical multivariate analysis of different structural goodness properties in order to characterize several detectable community topologies. Specifically, we show that a combination of two representative structural dimensions including community transitivity and hub dominance allows to distinguish different topologies such as star-based, clique-based, string-based and grid-based structures. Additionally, these classes of topology disclose structural proximities with those of graphs created by Erd\H{o}s-R\'{e}nyi, Watts-Strogatz and Barab\'{a}si-Albert generative models. We illustrate popular community topologies identified by different detection methods on a large dataset composing many network categories and associate their structures with the most related graph generative model. Interestingly, this conjunctive representation sheds light on fundamental differences between mesoscopic structures in various network categories including: communication, information, biological, technological, social, ecological, synthetic networks and more.
%G English
%L hal-01809064
%U https://hal-imt-atlantique.archives-ouvertes.fr/hal-01809064
%~ CNRS
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ ENIB
%~ LAB-STICC_ENIB
%~ LAB-STICC
%~ LAB-STICC_TB
%~ UNIV-BREST
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMTA_LUSSI
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Chapitre d'ouvrage
Bothorel Cécile, Lathia Neal, Picot-Clemente Romain, Noulas Anastasios
Location Recommendation with Social Media Data
Social Information Access, Volume 10100, Springer, Cham, pp.624 - 653, 2018, Lecture Notes in Computer Science book series (LNCS), 978-3-319-90091-9. ⟨10.1007/978-3-319-90092-6_16⟩
Details
Quote
Bibtext :
@incollection{bothorel:hal-01794923,
TITLE = {{Location Recommendation with Social Media Data}},
AUTHOR = {BOTHOREL, C{\'e}cile and Lathia, Neal and Picot-Clemente, Romain and Noulas, Anastasios},
URL = {https://hal.archives-ouvertes.fr/hal-01794923},
BOOKTITLE = {{Social Information Access}},
HAL_LOCAL_REFERENCE = {18919},
PUBLISHER = {{Springer, Cham}},
SERIES = { Lecture Notes in Computer Science book series (LNCS)},
VOLUME = { Volume 10100},
PAGES = {624 - 653},
YEAR = {2018},
DOI = {10.1007/978-3-319-90092-6\_16},
KEYWORDS = {Location data ; Venue data ; Location-based social networks ; Points-of-Interest (POIs) ; Location recommendation ; Recommendation using location data ; New places recommendation ; Events recommendation ; The next place to go recommendation ; Neighbourhoods recommendation ; Place search},
HAL_ID = {hal-01794923},
HAL_VERSION = {v1},
}
Endnote :
%0 Book Section
%T Location Recommendation with Social Media Data
%+ Lab-STICC_IMTA_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%+ Skyscanner (Skyscanner)
%+ Saagie (Saagie)
%+ School of Computing and Communications (University of Lancaster)
%A BOTHOREL, Cécile
%A Lathia, Neal
%A Picot-Clemente, Romain
%A Noulas, Anastasios
%@ 978-3-319-90091-9
%Z 18919
%B Social Information Access
%I Springer, Cham
%S Lecture Notes in Computer Science book series (LNCS)
%V Volume 10100
%P 624 - 653
%8 2018
%D 2018
%R 10.1007/978-3-319-90092-6_16
%K Location data
%K Venue data
%K Location-based social networks
%K Points-of-Interest (POIs)
%K Location recommendation
%K Recommendation using location data
%K New places recommendation
%K Events recommendation
%K The next place to go recommendation
%K Neighbourhoods recommendation
%K Place search
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Machine Learning [cs.LG]
%Z Computer Science [cs]/Mobile Computing
%Z Computer Science [cs]/Computers and Society [cs.CY]
%Z Computer Science [cs]/Information Retrieval [cs.IR]Book sections
%X Smartphones with inbuilt location-sensing technologies are now creating a new realm for recommender systems research and pratice. In this chapter, we focus on recommender systems that use location data to help users navigate the physical world. We examine various recommendation problems: recommending new places, recommending the next place to visit, events to attend, and recommending neighbourhoods or large areas to explore further. Lastly, we discuss how (personalized) place search is analogous to web search. For each of these domains, we present relevant data, algorithms, and methods, and we illustrate how researchers are investigating them with examples from the literature. We close by summarizing key aspects and suggesting future directions.
%G English
%L hal-01794923
%U https://hal.archives-ouvertes.fr/hal-01794923
%~ CNRS
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ ENIB
%~ TDS-MACS
%~ LAB-STICC
%~ UNIV-BREST
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMTA_LUSSI
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Chapitre d'ouvrage
Billot Romain, Bothorel Cécile, Lenca Philippe
Introduction to Big Data and Its Applications in Insurance
Big Data for Insurance Companies, Big Data, Artificial Intelligence and Data Analysis, 1, ISTE Editions, Wiley, pp.1 - 25, 2018, Innovation, Entrepreneurship and Management, 9781786300737
Details
Quote
Bibtext :
@incollection{billot:hal-01686059,
TITLE = {{Introduction to Big Data and Its Applications in Insurance}},
AUTHOR = {Billot, Romain and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01686059},
BOOKTITLE = {{Big Data for Insurance Companies}},
HAL_LOCAL_REFERENCE = {18487},
PUBLISHER = {{ISTE Editions, Wiley}},
SERIES = {Innovation, Entrepreneurship and Management},
VOLUME = {Big Data, Artificial Intelligence and Data Analysis, 1},
PAGES = {1 - 25},
YEAR = {2018},
KEYWORDS = {Big data ; Insurance industry},
HAL_ID = {hal-01686059},
HAL_VERSION = {v1},
}
Endnote :
%0 Book Section
%T Introduction to Big Data and Its Applications in Insurance
%+ Lab-STICC_IMTA_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%A Billot, Romain
%A BOTHOREL, Cécile
%A Lenca, Philippe
%@ 9781786300737
%Z 18487
%B Big Data for Insurance Companies
%I ISTE Editions, Wiley
%S Innovation, Entrepreneurship and Management
%V Big Data, Artificial Intelligence and Data Analysis, 1
%P 1 - 25
%8 2018
%D 2018
%K Big data
%K Insurance industry
%Z Computer Science [cs]/Hardware Architecture [cs.AR]
%Z Computer Science [cs]/Databases [cs.DB]
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]
%Z Computer Science [cs]/Computers and Society [cs.CY]
%Z Statistics [stat]/Applications [stat.AP]
%Z Statistics [stat]/Machine Learning [stat.ML]Book sections
%X We present an introduction to big data and its application in insurance (impact of big data, 5V and others, tools and architecture for big data, example of application in insurance).
%G English
%L hal-01686059
%U https://hal.archives-ouvertes.fr/hal-01686059
%~ INSTITUT-TELECOM
%~ TELECOM-BRETAGNE
%~ CNRS
%~ UNIV-UBS
%~ ENIB
%~ TDS-MACS
%~ LAB-STICC
%~ UNIV-BREST
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Communication dans un congrès
Dao Vinh-Loc, Bothorel Cécile, Lenca Philippe
Community detection methods can discover better structural clusters than ground-truth communities
2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, Jul 2017, Sydney, Australia. ⟨10.1145/3110025.3110053⟩
Bibtext :
@inproceedings{dao:hal-01577343,
TITLE = {{Community detection methods can discover better structural clusters than ground-truth communities}},
AUTHOR = {Dao, Vinh-Loc and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01577343},
BOOKTITLE = {{2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining}},
ADDRESS = {Sydney, Australia},
HAL_LOCAL_REFERENCE = {18122},
YEAR = {2017},
MONTH = Jul,
DOI = {10.1145/3110025.3110053},
KEYWORDS = {Modular structures ; Community detection in graphs ; Community evaluation metrics ; Network science ; Complex network analysis ; Graph clustering},
PDF = {https://hal.archives-ouvertes.fr/hal-01577343/file/CRV_Webversion.pdf},
HAL_ID = {hal-01577343},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Community detection methods can discover better structural clusters than ground-truth communities
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%+ Lab-STICC_IMTA_CID_DECIDE
%A Dao, Vinh-Loc
%A BOTHOREL, Cécile
%A Lenca, Philippe
%< avec comité de lecture
%Z 18122
%( 2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
%B 2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
%C Sydney, Australia
%8 2017-07-31
%D 2017
%R 10.1145/3110025.3110053
%K Modular structures
%K Community detection in graphs
%K Community evaluation metrics
%K Network science
%K Complex network analysis
%K Graph clustering
%Z Computer Science [cs]
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Physics [physics]/Physics [physics]/Data Analysis, Statistics and Probability [physics.data-an]
%Z Computer Science [cs]/Social and Information Networks [cs.SI]
%Z Computer Science [cs]/Numerical Analysis [cs.NA]Conference papers
%X Community detection emerged as an important exploratory task in complex networks analysis across many scientific domains. Many methods have been proposed to solve this problem, each one with its own mechanism and sometimes with a different notion of community. In this article, we bring most common methods in the literature together in a comparative approach and reveal their performances in both real-world networks and synthetic networks. Surprisingly, many of those methods discovered better communities than the declared ground-truth communities in terms of some topological goodness features, even on benchmarking networks with built-in communities. We illustrate different structural characteristics that these methods could identify in order to support users to choose an appropriate method according to their specific requirements on different structural qualities.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01577343/document
%2 https://hal.archives-ouvertes.fr/hal-01577343/file/CRV_Webversion.pdf
%L hal-01577343
%U https://hal.archives-ouvertes.fr/hal-01577343
%~ CNRS
%~ UNIV-UBS
%~ INSTITUT-TELECOM
%~ ENIB
%~ LAB-STICC
%~ UNIV-BREST
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Communication dans un congrès
Dao Vinh-Loc, Bothorel Cécile, Lenca Philippe
Community structures evaluation in complex networks: A descriptive approach
NetSci-X 2017 : International School and Conference on Network Science, Jan 2017, Tel Aviv, Israel. pp.11-19, ⟨10.1007/978-3-319-55471-6_2⟩
Bibtext :
@inproceedings{dao:hal-01513246,
TITLE = {{Community structures evaluation in complex networks: A descriptive approach}},
AUTHOR = {Dao, Vinh-Loc and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01513246},
BOOKTITLE = {{NetSci-X 2017 : International School and Conference on Network Science}},
ADDRESS = {Tel Aviv, Israel},
HAL_LOCAL_REFERENCE = {17743},
SERIES = {3rd International Winter School and Conference on Network Science (NetSci-X 2017)},
PAGES = {11-19},
YEAR = {2017},
MONTH = Jan,
DOI = {10.1007/978-3-319-55471-6\_2},
KEYWORDS = {Network Science ; Community Detection ; Community Structure ; Network Partition ; Understanding of modular structure ; Descriptive KPI},
PDF = {https://hal.archives-ouvertes.fr/hal-01513246/file/294-NetSciX2017-5.pdf},
HAL_ID = {hal-01513246},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Community structures evaluation in complex networks: A descriptive approach
%+ Lab-STICC_TB_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (LUSSI)
%A Dao, Vinh-Loc
%A BOTHOREL, Cécile
%A Lenca, Philippe
%< avec comité de lecture
%Z 17743
%( Springer Proceedings in Complexity
%B NetSci-X 2017 : International School and Conference on Network Science
%C Tel Aviv, Israel
%3 3rd International Winter School and Conference on Network Science (NetSci-X 2017)
%P 11-19
%8 2017-01-15
%D 2017
%R 10.1007/978-3-319-55471-6_2
%K Network Science
%K Community Detection
%K Community Structure
%K Network Partition
%K Understanding of modular structure
%K Descriptive KPI
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Machine Learning [cs.LG]
%Z Computer Science [cs]/Computers and Society [cs.CY]Conference papers
%X Evaluating a network partition just only via conventional quality metrics - such as modularity, conductance or normalized mutual of information - is usually insufficient. Indeed, global quality scores of a network partition or its clusters do not provide many ideas about their structural characteristics. Furthermore, quality metrics often fail to reach an agreement especially in networks whose modular structures are not very obvious. Evaluating the goodness of network partitions in function of desired structural properties is still a challenge. Here, we propose a methodology that allows one to expose structural information of clusters in a network partition in a comprehensive way, thus eventually helps one to compare communities identified by different community detection methods. This descriptive approach also helps to clarify the composition of communities in real-world networks. The methodology hence bring us a step closer to the understanding of modular structures in complex networks.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01513246/document
%2 https://hal.archives-ouvertes.fr/hal-01513246/file/294-NetSciX2017-5.pdf
%L hal-01513246
%U https://hal.archives-ouvertes.fr/hal-01513246
%~ TELECOM-BRETAGNE
%~ INSTITUT-TELECOM
%~ CNRS
%~ UNIV-UBS
%~ ENIB
%~ LAB-STICC_ENIB
%~ TDS-MACS
%~ LAB-STICC
%~ LAB-STICC_TB
%~ UNIV-BREST
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Chapitre d'ouvrage
Billot Romain, Bothorel Cécile, Lenca Philippe
Introduction au big data et application à l'assurance
Le big data pour les compagnies d'assurance, 1, ISTE Editions, pp.17 - 39, 2017, Innovation, entrepreneuriat et gestion, Série Big Data, IA et analyse de données, 978-1-78405-298-0
Details
Quote
Bibtext :
@incollection{billot:hal-01596095,
TITLE = {{Introduction au big data et application {\`a} l'assurance}},
AUTHOR = {Billot, Romain and BOTHOREL, C{\'e}cile and Lenca, Philippe},
URL = {https://hal.archives-ouvertes.fr/hal-01596095},
BOOKTITLE = {{Le big data pour les compagnies d'assurance}},
HAL_LOCAL_REFERENCE = {18066},
PUBLISHER = {{ISTE Editions}},
SERIES = {Innovation, entrepreneuriat et gestion, S{\'e}rie Big Data, IA et analyse de donn{\'e}es},
VOLUME = {1},
PAGES = {17 - 39},
YEAR = {2017},
KEYWORDS = {Big data ; M{\'e}gadonn{\'e}es ; Informatique d{\'e}cisionnelle ; Business inteligence ; Hadoop ; Assurance},
HAL_ID = {hal-01596095},
HAL_VERSION = {v1},
}
Endnote :
%0 Book Section
%T Introduction au big data et application à l'assurance
%+ Lab-STICC_TB_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (IMT Atlantique - LUSSI)
%+ Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance (Lab-STICC)
%A Billot, Romain
%A BOTHOREL, Cécile
%A Lenca, Philippe
%@ 978-1-78405-298-0
%Z 18066
%B Le big data pour les compagnies d'assurance
%I ISTE Editions
%S Innovation, entrepreneuriat et gestion, Série Big Data, IA et analyse de données
%V 1
%P 17 - 39
%8 2017
%D 2017
%K Big data
%K Mégadonnées
%K Informatique décisionnelle
%K Business inteligence
%K Hadoop
%K Assurance
%Z Computer Science [cs]/Databases [cs.DB]
%Z Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]
%Z Computer Science [cs]/Computers and Society [cs.CY]
%Z Statistics [stat]/Applications [stat.AP]
%Z Statistics [stat]/Machine Learning [stat.ML]Book sections
%X On présente une introduction au big data et son application à l'assurance (impact des mégadonnées, les 5V et d'autres encore, outils et architecture pour les mégadonnées, exemple d'application à l'assurance).
%G French
%L hal-01596095
%U https://hal.archives-ouvertes.fr/hal-01596095
%~ INSTITUT-TELECOM
%~ TELECOM-BRETAGNE
%~ CNRS
%~ UNIV-UBS
%~ ENIB
%~ LAB-STICC_ENIB
%~ TDS-MACS
%~ LAB-STICC
%~ LAB-STICC_TB
%~ UNIV-BREST
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Article dans une revue
Bothorel Cécile, Cruz Gomez Juan David, Matteo Magnani, Micenkova Barbora
Clustering attributed graphs: models, measures and methods
Network Science, Cambridge Journals, 2015, 3 (03), pp.408 - 444. ⟨10.1017/nws.2015.9⟩
Details
Quote
Bibtext :
@article{bothorel:hal-01257833,
TITLE = {{Clustering attributed graphs: models, measures and methods}},
AUTHOR = {BOTHOREL, C{\'e}cile and CRUZ GOMEZ, Juan David and Matteo, Magnani and Micenkova, Barbora},
URL = {https://hal.archives-ouvertes.fr/hal-01257833},
JOURNAL = {{Network Science}},
HAL_LOCAL_REFERENCE = {15007},
PUBLISHER = {{Cambridge Journals}},
VOLUME = {3},
NUMBER = {03},
PAGES = {408 - 444},
YEAR = {2015},
MONTH = Sep,
DOI = {10.1017/nws.2015.9},
KEYWORDS = {Attributed edges ; Attributed nodes ; Clustering attributed graphs ; Survey ; Models ; Measures ; Methods},
HAL_ID = {hal-01257833},
HAL_VERSION = {v1},
}
Endnote :
%0 Journal Article
%T Clustering attributed graphs: models, measures and methods
%+ Lab-STICC_TB_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (LUSSI)
%+ Computing Science Division (Uppsala University)
%+ Data Intensive Systems, Department of Computer Science (Aarhus University)
%A BOTHOREL, Cécile
%A CRUZ GOMEZ, Juan David
%A Matteo, Magnani
%A Micenkova, Barbora
%< avec comité de lecture
%Z 15007
%@ 2050-1242
%J Network Science
%I Cambridge Journals
%V 3
%N 03
%P 408 - 444
%8 2015-09
%D 2015
%R 10.1017/nws.2015.9
%K Attributed edges
%K Attributed nodes
%K Clustering attributed graphs
%K Survey
%K Models
%K Measures
%K Methods
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Machine Learning [cs.LG]Journal articles
%X Clustering a graph, i.e., assigning its nodes to groups, is an important operation whose best known application is the discovery of communities in social networks. Graph clustering and community detection have traditionally focused on graphs without attributes, with the notable exception of edge weights. However, these models only provide a partial representation of real social systems, that are thus often described using node attributes, representing features of the actors, and edge attributes, representing different kinds of relationships among them. We refer to these models as attributed graphs. Consequently, existing graph clustering methods have been recently extended to deal with node and edge attributes. This article is a literature survey on this topic, organizing and presenting recent research results in a uniform way, characterizing the main existing clustering methods and highlighting their conceptual differences. We also cover the important topic of clustering evaluation and identify current open problems.
%G English
%L hal-01257833
%U https://hal.archives-ouvertes.fr/hal-01257833
%~ INSTITUT-TELECOM
%~ TELECOM-BRETAGNE
%~ CNRS
%~ UNIV-BREST
%~ UNIV-UBS
%~ LAB-STICC_ENIB
%~ ENIB
%~ LAB-STICC
%~ LAB-STICC_TB
%~ IMTA_LUSSI
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
Communication dans un congrès
Picot Clemente Romain, Bothorel Cécile, Jullien Nicolas
Social Interactions vs Revisions, What Is Important for Promotion in Wikipedia?
ASONAM 2015 : IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining , Aug 2015, Paris, France. pp.888 - 893, ⟨10.1145/2808797.2810063⟩
Details
Quote
Bibtext :
@inproceedings{picotclemente:hal-01243183,
TITLE = {{Social Interactions vs Revisions, What Is Important for Promotion in Wikipedia?}},
AUTHOR = {PICOT CLEMENTE, Romain and BOTHOREL, C{\'e}cile and JULLIEN, Nicolas},
URL = {https://hal.archives-ouvertes.fr/hal-01243183},
BOOKTITLE = {{ASONAM 2015 : IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining }},
ADDRESS = {Paris, France},
HAL_LOCAL_REFERENCE = {15568},
PAGES = {888 - 893},
YEAR = {2015},
MONTH = Aug,
DOI = {10.1145/2808797.2810063},
KEYWORDS = {Epistemic Community ; Random forest ; Wikipedia ; Request for adminship ; Promotion ; Predictive model},
HAL_ID = {hal-01243183},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Social Interactions vs Revisions, What Is Important for Promotion in Wikipedia?
%+ Lab-STICC_TB_CID_DECIDE
%+ Département Logique des Usages, Sciences sociales et Sciences de l'Information (LUSSI)
%+ Môle Armoricain de Recherche sur la SOciété de l'information et des usages d'INternet (MARSOUIN)
%+ Laboratoire Information, Coordination, Incitations (ICI)
%A PICOT CLEMENTE, Romain
%A BOTHOREL, Cécile
%A JULLIEN, Nicolas
%< avec comité de lecture
%Z 15568
%B ASONAM 2015 : IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
%C Paris, France
%P 888 - 893
%8 2015-08-25
%D 2015
%R 10.1145/2808797.2810063
%K Epistemic Community
%K Random forest
%K Wikipedia
%K Request for adminship
%K Promotion
%K Predictive model
%Z Computer Science [cs]/Data Structures and Algorithms [cs.DS]
%Z Computer Science [cs]/Machine Learning [cs.LG]
%Z Humanities and Social Sciences/Economies and financesConference papers
%X In epistemic communities, people are said to be selected on their contribution in knowledge to the project (articles, codes, etc.). However, the socialization process is an important factor for inclusion, sustainability as a contributor, and promotion. Finally, what matters for being promoted? Being a good contributor? Being a good animator? Knowing the boss? We explore this question by looking at the election process for administrators in the English Wikipedia. We used the candidates' revisions and/or social attributes to construct a predictive model of promotion success, based on the candidates' past behavior and a random forest algorithm. Our model explains 78% of the results, which is better than the former models. It also helps to refine the explanation of the election process.
%G English
%L hal-01243183
%U https://hal.archives-ouvertes.fr/hal-01243183
%~ INSTITUT-TELECOM
%~ TELECOM-BRETAGNE
%~ UR2-HB
%~ CNRS
%~ UNIV-BREST
%~ UNIV-UBS
%~ UNIV-RENNES1
%~ AO-ECONOMIE
%~ ENIB
%~ LAB-STICC_ENIB
%~ LAB-STICC
%~ SHS
%~ UR1-HAL
%~ LAB-STICC_TB
%~ UNIV-RENNES2
%~ TEST-UNIV-RENNES
%~ TEST-UR-CSS
%~ IMTA_LUSSI
%~ LEGO
%~ UNIV-RENNES
%~ LAB-STICC_IMTA_CID_DECIDE
%~ IMT-ATLANTIQUE
%~ LAB-STICC_IMTA
%~ MARSOUIN-IMTA
%~ IBSHS
See all publications