Alexandre Termier - Pattern mining rock: more, faster, better

Organized by: 

Alexandre Termier


- Location

  • Grenoble University main campus
  • UFR IM2AG building F, room F018.
  • Address : 60 rue de la Chimie, Saint Martin d’Hères

- Jury

  • President : Claudia Roncancio, Pr, Grenoble INP
  • Reviewer : Hiroki Arimura, Pr, Hokkaido University, Japan
  • Reviewer : Jean-François Boulicaut, Pr, INSA Lyon
  • Reviewer : Mohammed J. Zaki, Rensselaer Polytechnic Institute, NY, USA
  • Examiner : Marie-Odile Cordier, Pr, Université Rennes 1
  • Examiner : Jean-François Méhaut, Pr, Université Joseph Fourier
  • Examiner : Marie-Christine Rousset, Pr, Université Joseph Fourier
  • Examiner : Takashi Washio, Osaka University, Japan

- See also :

Pattern mining is the area of data mining concerned with finding regularities in data. In this defense I will present my contributions to this domain along three axes :

  1. The domain is young, and there are still some kinds of regularities that data analysts would like to discover in data but that are not handled. We contributed two new pattern definitions extending the reach of data analysis by pattern mining : gradual patterns and periodic patterns with unrestricted gaps. We also proposed ParaMiner, a pioneering algorithm for generic pattern mining, allowing practitioners to directly specify the patterns they are interested in.
  2. Pattern mining is extremely demanding on computational resources. In order to reduce the mining time, we studied how to exploit the parallelism of multicore processors. Our results show that some well established techniques in pattern mining are ill-adapted for parallelism, and propose solutions.
  3. Our ultimate goal is to make pattern mining easier to use by data analysts. There is a lot to do in this area, as currently they are presented with unusable lists of millions of patterns. We will present our first results in the context of mining execution traces of processors.