Hide menu

732A31 Data Mining - Clustering and Association Analysis

Syllabus


syllabus

AIM OF THE COURSE

The course lays the foundation for professional work and research in which large amounts of data are explored, modified, modelled and assessed to uncover previously unknown patterns and trends. The course focuses on clustering and association analysis.

Having completed the course, the student should be able to:

  • understand and be able to use important terminology in data mining
  • understand and use the theory behind clustering and association analysis
  • use knowledge about techniques for clustering and association analysis
  • demonstrate insightful assessment of the quality of given data sets and the information content on which clustering and association analysis can be based
  • use and evaluate tools for clustering and association analysis
  • design and perform a data analysis task using clustering and association analysis for a given data mining problem and evaluate the results

CONTENTS

Association analysis: concepts and methods related to frequent item sets and association rules such as Apriori principle, FP-growth, evaluation of association rules,

Clustering: concepts and methods related to partitional clustering methods (e.g. K-means), hierarchical clustering methods, density-based clustering methods (e.g. DBSCAN), cluster evaluation, outlier analysis

TEACHING

The teaching comprises lectures, seminars, and computer laboratory and project work. Lectures are devoted to theory, concepts and techniques. The techniques are practised in the computer laboratory and project work. The seminars comprise student presentations and discussions of assignments. Language of instruction: English.

EXAMINATION

- written examination
- presentation, laboratory and project work

Page responsible: Patrick Lambrix
Last updated: 2011-03-29