Classification Days

Data Mining consists in extracting previously unknown, potentially useful and reliable patterns from a given set of data. The use of classification techniques, as well as the use of clustering techniques, can provide important information for applications arising in various disciplines. The aim of our Classification Days is to bring together researchers either working on the development of specific data mining techniques, or interested in using such techniques for extracting useful information from databases obtained when studying particular real-life problems.

During each CD, speakers are invited to give presentations of their scientific results and of their scientific projects; these presentations will then be used as a starting point for a deeper discussion involving all partecipants. At each CD, previous discussions are reminded and, when this is the case, new collaborations issued from previous CDs are presented.

This page is for collecting all CD's programs, discussions and slides provided by the speakers. It also collects the main references at the basis of the discussions, with a direct link to the original publication.

Important: PhD students of the Matisse Doctoral School, with the agreement of their supervisors, can have their attendance to the CDs validated as "heures de formation".

CD4 "Deep Learning", October 17th, 2017.
  • Yannis Avrithis, INRIA Rennes
    Deep Learning, Unsupervised Learning and Image Retrieval

  • Ludovic Hoyet, INRIA Rennes
    On the Classification of Human Movements

CD3 "Time-series classification", June 28th, 2017.
CD2 "Enhancing k-means", February 7th, 2017.
CD1 "Supervised biclustering", November 10th, 2016.


External links


The CDs are organized by Antonio Mucherino and the main participants are the members of the MimeTIC team at IRISA/INRIA. However, the speakers from other teams are regurarly invited to give presentations and take part to the discussions.


To INRIA for providing the coffee breaks!

Back Home