Nowadays, data drives the world. Public and private organisations in all sectors face an avalanche of digital data. While at first glance this appears to be favourable for our knowledge-based society, in many ways it is a burden. Data is of great value only after it has been refined and analysed, in order to produce information and knowledge that can be used for decision making in a particular domain of interest. Usage is no longer restricted to simple querying or data mining, it increasingly requires exploration, recommendation, explanation, and visualisation, along with special treatment of time and location.
This specialization disseminates essential knowledge about the ethical generation, collection, and exploitation of data for efficient decision-making systems. It focuses on models, algorithms, and technologies related to massive data analytics. It covers theoretical foundations such as decision-making in uncertain situations, machine learning methods, graph management and analytics as well as visualisation and innovation.
This specialization is underpinned by the Erasmus Mundus BDMA (Big Data Management and Analytics) Master’s program.
Lieu(x) d'enseignement
GIF SUR YVETTE
Pré-requis, profil d’entrée permettant d'intégrer la formation
Science and engineering graduates, holding diploma equivalent to high school diploma + 4 years of higher education studies. Including Erasmus Mundus BDMA programme students (achieving ULB and UPC BDMA programme).
Compétences
The students acquire fundamental knowledge related to models, algorithms, and technologies for decision-support systems.
The students acquire fundamental knowledge in data management and data analytics that any data specialist must have.
The students develop fundamental technical competences as well as skills in research, ethics and entrepreneurship.
Profil de sortie des étudiants ayant suivi la formation
Digital Transformation Leader, Data Scientist, System Analyst, Data Specialist, and Data Administrator
Débouchés de la formation
Career opportunities for students in this field include executive IT positions in industry or the services sector, or data management research and development positions in universities or public and private research organizations, or in major groups and startups. More specifically, career opportunities following the BDMA track include those linked to models, algorithms, and technologies related to decision-support systems and massive data analytics.
Collaboration(s)
Laboratoire(s) partenaire(s) de la formation
Laboratoire de recherche en informatique
Laboratoire des Signaux et Systèmes
Centre de Vision Numérique
Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur.
Programme
Le parcours BDMA est uniquement ouvert en M2. Il s'adresse aux étudiants inscrits au programme Erasmus Mundus. La validation du premier semestre consiste à valider 11 UEs.
This course aims at preparing the students for the master thesis of the 4th semester. The students will learn how to manage a research project related to massive and heterogeneous data management and analytics from scratch, working in a team, and using all the steps required in a scientific methodology. During this course the students will attend seminars in order to have a better understanding of research methodologies and to be aware of some ongoing research projects presented by researchers.
Prérequis :
Upon successful completion of this course, the student is able to manage a scientific project from scratch in a team and provide scale-up algorithms and a program prototype for massive data management and analysis.
Bibliographie :
Scientific papers will be distributed by the lecturer and the invited speakers according to the covered topics.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
This course aims at presenting classical decision models with a special emphasis on decision making in uncertain situations, decision with multiple attribute, and decision with multiple stakeholders.
During the course, various applications will be presented, emphasizing the practical interest and applicability of the models in real-world decision situations.
Learning outcomes:
Upon successful completion of this course, the student will acquire knowledge and skills about:
decision models, validity of the decision models
the three levels of decision analysis: representation of observed decision behavior (descriptive analysis), decision aiding and recommendation (prescriptive analysis), and the design of artificial decision agents (normative analysis).
Prérequis :
Operational research algorithms foundations.
Bibliographie :
Denis Bouyssou, Thierry Marchant, Marc Pirlot, Alexis Tsoukias, Philippe Vincke, Evaluation and decision models with multiple criteria: Stepping stones for the analyst, Springer, International Series in Operations Research and Management Science Volume 86, 2006.
William W. Cooper, Lawrence M. Seiford, and Kaoru Tone, Introduction to Data Envelopment Analysis and Its Uses, Springer, 2006.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
This course aims at presenting classical decision models with a special emphasis on decision making in uncertain situations, decision with multiple attribute, and decision with multiple stakeholders.
During the course, various applications will be presented, emphasizing the practical interest and applicability of the models in real-world decision situations.
Learning outcomes:
Upon successful completion of this course, the student will acquire knowledge and skills about:
decision models, validity of the decision models
the three levels of decision analysis: representation of observed decision behavior (descriptive analysis), decision aiding and recommendation (prescriptive analysis), and the design of artificial decision agents (normative analysis).
Prérequis :
Operational research algorithms foundations.
Bibliographie :
Denis Bouyssou, Thierry Marchant, Marc Pirlot, Alexis Tsoukias, Philippe Vincke, Evaluation and decision models with multiple criteria: Stepping stones for the analyst, Springer, International Series in Operations Research and Management Science Volume 86, 2006.
William W. Cooper, Lawrence M. Seiford, and Kaoru Tone, Introduction to Data Envelopment Analysis and Its Uses, Springer, 2006.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
The objectives of this course are to provide the student: (i) knowledge about intellectual and industrial properties, data protection and security in European research context, (ii) an overview about current and innovative company projects and technology needs for real data analytics and machine learning.
Upon completing the course, the student will acquire knowledge and skills about:
intellectual and industrial properties, data protection in European research context;
corporate and entrepreneurship culture;
innovative projects and technologies related to massive and real data management, analytics and machine learning.
Bibliographie :
Scientific papers will be distributed by the course lecturer and invited speakers according to the topics covered.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
The goal of this course is to provide the student with knowledge about supervised, unsupervised, deep and reinforcement learning paradigms; the mathematical foundations and practices of different variants of machine learning methods.
Upon successful completion of this course, the student will be able to:
choose the best techniques to solve a given machine learning task;
tune the parameters of the chosen method;
interpret the results and compare different learning methods.
Prérequis :
Probability and Statistics
Data Mining foundations.
Bibliographie :
David J. Hand, Heikki Mannila and Padhraic Smyth, Principles of Data Mining, The MIT Press, 2001.
Trevor Hastie, Robert Tibshirani, Jerome Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition, Springer, 2009.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
The goal of this course is to provide the student with knowledge about supervised, unsupervised, deep and reinforcement learning paradigms; the mathematical foundations and practices of different variants of machine learning methods.
Upon successful completion of this course, the student will be able to:
choose the best techniques to solve a given machine learning task;
tune the parameters of the chosen method;
interpret the results and compare different learning methods.
Prérequis :
Probability and Statistics
Data mining foundations.
Bibliographie :
David J. Hand, Heikki Mannila and Padhraic Smyth, Principles of Data Mining, The MIT Press, 2001.
Trevor Hastie, Robert Tibshirani, Jerome Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition, Springer, 2009.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
The objectives of this course is to provide the student with
knowledge about designing high-performance and scalable algorithms for massive graph analytics. The course focuses on modeling and querying massive graph data in a distributed environment, designing algorithms, complexity analysis and optimization, for massive data graph problem analytics.
Upon successful completion of this course, the student is able to:
model and query massive graph data in a distributed environment
design and analyse efficient graph algorithms in real-world data-intensive applications;
develop efficient applications using the best practices in a distributed environment (Spark, MapReduce, Neo4J, GraphX, etc.).
Prérequis :
Graph theory foundations.
Bibliographie :
Richard Brath, David Jonker, Graph Analysis and Visualization: Discovering Business Opportunity in Linked Data, Wiley, 2015
Handbook Of Large-Scale Random Networks, Editors: Bollobas, Bela, Kozma, Robert, Miklós, Dezs? (Eds.), Springer, 2008.
Yan, Da, Tian, Yuanyuan, Cheng, James, Systems for Big Graph Analytics, Springer 2017.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
The objectives of this course is to provide the student with
knowledge about designing high-performance and scalable algorithms for massive graph analytics. The course focuses on modeling and querying massive graph data in a distributed environment, designing algorithms, complexity analysis and optimization, for massive data graph problem analytics.
Upon successful completion of this course, the student is able to:
model and query massive graph data in a distributed environment
design and analyse efficient graph algorithms in real-world data-intensive applications;
develop efficient applications using the best practices in a distributed environment (Spark, MapReduce, Neo4J, GraphX, etc.).
Prérequis :
Graph theory foundations.
Bibliographie :
Richard Brath, David Jonker, Graph Analysis and Visualization: Discovering Business Opportunity in Linked Data, Wiley, 2015
Handbook Of Large-Scale Random Networks, Editors: Bollobas, Bela, Kozma, Robert, Miklós, Dezs? (Eds.), Springer, 2008.
Yan, Da, Tian, Yuanyuan, Cheng, James, Systems for Big Graph Analytics, Springer 2017.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
In this course students learn how to bring together automated and human-driven data analysis approaches; including innovative aspects such as exploratory data analysis, perception and cognition, storytelling, text analysis, and multi-dimensional data representation.
Upon completing the course, the student will be able to:
understand basic concepts, theories, and methodologies of Visual Analytics;
analyse data using appropriate visual analytics thinking and techniques;
present data using appropriate visual communication and graphical methods;
design and implement a Visual Analytics system for supporting decision making.
Prérequis :
Database systems
Data mining foundations.
Bibliographie :
Edward Tufte, Envisioning Information, Graphics Press, 1990.
Robert Spence, Information Visualisation: Design for Interaction, Second Edition, Prentice Hall, 2007.
Colin Ware, Information Visualisation: Perception for Design, Second Edition, Morgan Kaufmann, 2004.
Big Data and Visual Analytics by Sang C. Suh, Thomas Anthony (Editors) 1st Edition Springer 2017.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
In this course students learn how to bring together automated and human-driven data analysis approaches; including innovative aspects such as exploratory data analysis, perception and cognition, storytelling, text analysis, and multi-dimensional data representation.
Upon completing the course, the student will be able to:
understand basic concepts, theories, and methodologies of Visual Analytics;
analyse data using appropriate visual analytics thinking and techniques;
present data using appropriate visual communication and graphical methods;
design and implement a Visual Analytics system for supporting decision making.
Prérequis :
Visual Analytics I.
Bibliographie :
Edward Tufte, Envisioning Information, Graphics Press, 1990.
Robert Spence, Information Visualisation: Design for Interaction, Second Edition, Prentice Hall, 2007.
Colin Ware, Information Visualisation: Perception for Design, Second Edition, Morgan Kaufmann, 2004.
Big Data and Visual Analytics by Sang C. Suh, Thomas Anthony (Editors) 1st Edition Springer 2017.
Période(s) et lieu(x) d’enseignement :
Période(s) :
Septembre - Octobre - Novembre - Décembre - Janvier - Février.
Lieu(x) :
GIF-SUR-YVETTE
Le deuxième semestre du M2 BDMA consiste en un stage long (5 à 6 mois) permettant d'obtenir 30 ECTS.
Fiche de choix de M2 (obligatoire pour les candidats inscrits en M1 à l'Université Paris-Saclay) à télécharger sur https://urlz.fr/i3Lo.
Document justificatif des candidats exilés ayant un statut de réfugié, protection subsidiaire ou protection temporaire en France ou à l’étranger (facultatif mais recommandé, un seul document à fournir) :
- Carte de séjour mention réfugié du pays du premier asile
- OU récépissé mention réfugié du pays du premier asile
- OU document du Haut Commissariat des Nations unies pour les réfugiés reconnaissant le statut de réfugié
- OU récépissé mention réfugié délivré en France
- OU carte de séjour avec mention réfugié délivré en France
- OU document faisant état du statut de bénéficiaire de la protection subsidiaire en France ou à l’étranger.
