New website

The master has a new home:

This page is not updated anymore….

Contributing Higher Education Institutions:

  • Ecole Polytechnique
  • Télécom ParisTech
  • ENSAE ParisTech
  • ENS Cachan
  • Université ParisSud


The Data Science Game 2015 was held in Paris and was sponsored by Google and Gapgemini. The competition lasted 2 days in a castle near Paris and 20 teams from different countries and various well-known universities (including Imperial College London, University of Amsterdam, ENS Cachan, Moscow State University, Telecom) took part in the challenge. The task was to classify YouTube videos into 15 categories using features such as title, description, topics, likes, duration, etc. The final ranking of the participating teams was based on the classification accuracy performance on an unknown test set given 30 minutes before the end of the competition.

The Polytechnique team (Poly Unicorns), consisting by the M2 – Data Science students G. Bekoulis, S. Rubrichi, M. Sakji and E. Tverdokhlebova participated in the challenge and ranked 7th among the participating universities. Their performance was 72.96% just 0.9% below the 2nd team. Congratulations!!


Experiments, observations, and numerical simulations in many areas of science and business are currently generating terabytes of data, and in some cases are on the verge of generating petabytes and beyond. Analyses of the information contained in these data sets have already led to major breakthroughs in fields ranging from genomics to astronomy and high energy physics and to the development of new information-based industries.

Traditional methods of analysis have been based largely on the assumption that analysts can work with data within the confines of their own computing environment, but the growth of “big data” is changing that paradigm, especially in cases in which massive amounts of data are distributed across locations.

Data mining of these massive data sets is transforming the way we think about crisis response, marketing, entertainment, cyber-security, and national intelligence. It is also transforming how we think about information storage and retrieval. Collections of documents, images, videos, and networks are being thought of not merely as bit strings to be stored, indexed, and retrieved, but also as potential sources of valuable information. Discovery and knowledge, requiring sophisticated analysis techniques that go far beyond classical indexing and keyword counting, aiming to find relational and semantic interpretations of the phenomena underlying the data.

Data Science and Bigdata are two key areas for positive interdisciplinary science involving mathematics, computer science. The context is the thorough presentation of state of the art mathematical and computational methods for the management and analysis of data of potentially very big scale.

A number of challenges in both data management and analysis require new approaches to support the big data era. These challenges span generation of the data, preparation for analysis, and policy-related challenges in its sharing and use, including the following:

  • Dealing with highly distributed data sources with parallel and distributed architectures
  • Tracking data provenance, from data generation through data preparation
  • Coping with sampling biases, different data formats and structures
  • Ensuring data integrity, security, sharing
  • Methods for massive data visualization
  • Learning from massive data and enabling predictions
  • Developing scalable and incremental algorithms for real-time analysis and decision-making.

The Math Bigdata Master aims to stimulate the demanding students with current and forward looking topics that integrate sound fundamental methods and practical applications in current and emerging domains.

The rationale

Data Science and Bigdata are two key areas for positive interdisciplinary science involving mathematics, computer science. The context is the management of heterogeneous data of potentially very big scale.