Unsupervised Learning

Course Code : IA341

Time Hours : 18 hours

Time Periods : not available yet

Lecturer : François-Xavier Jollois francois-xavier.jollois[at]parisdescartes.fr

Objective

The purpose of this course is to describe the unsupervised learning techniques most commonly used. Different types of data from several fields such as bioinformatics, e-commerce and text mining illustrate the interest of the various methods studied.

Acquired Skills

In which context each method can be used, how to interpret its results on real situations and how to select the one or several methods to be used.

Contents

  • Similarity, dissimilarity and distance
  • Hierarchical agglomerative clustering
  • Partitional clustering : family of the K-means
  • Spectral clustering
  • SOM (self-Organizing map)
  • Semi-supervised clustering
  • Knowledge-based clustering
  • Association rule learning

References

  • Clustering for data mining, Boris Mirkin, Chapman & Hall/CRC, (2005)
  • Data Analysis, Gérad Govaert, ISTE Ltd and John Wiley & Sons Inc (2009)
  • Data mining : Concepts and Techniques, Jiawei Han, Micheline Kamber, Morgan Kaufman publishers (2005)