An exploratory study of feature selection techniques for unsupervised learning

01 October 2007 → 30 September 2013
Regional and community funding: Special Research Fund, Research Foundation - Flanders (FWO)
Research disciplines
  • Natural sciences
    • Applied mathematics in specific fields
    • Artificial intelligence
    • Scientific computing
    • Bioinformatics and computational biology
  • Social sciences
    • Cognitive science and intelligent systems
  • Medical and health sciences
    • Bioinformatics and computational biology
    • Bioinformatics and computational biology
    • Public health care
    • Public health services
    • Bioinformatics and computational biology
  • Engineering and technology
    • Scientific computing
feature selection machine learning clustering
Project description

Feature selection is an important aspect of present day data mining research. The need for feature selection is currently increasing as gradually more large and high dimensional datasets are becoming available. In this research, we will focus on how the existing taxonomy of feature selection for classificationcan be transferred to the related problem of clustering (unsupervised learning).