63765 Data Mining

63765 Data Mining

  • Study programme and level: Interdisciplinary University Study Programme in Administrative Information Science - 1st Cycle
  • 6 ECTS
  • Course type: Elective
  • Lectures: 45
  • Tutorial: 30
  • Individual work: 105
  • Lecturer: Janez Demšar, PhD

 

 

1. Objectives and competences

The purpose of the course is to teach students how to mine data. After completing the course, the students should be able to use the data for extraction of patterns and hypothesis that should be potentially useful for the data owner.

2. Content

The course is divided into the following lectures:

  • Introduction, motivation
  • Refreshment of machine learning and statistics
  • Data visualization, good and bad examples
  • How to combine visualization, machine learning and statistics
  • Discretization of continuous data; handling unknown and noisy data using a combination of automated methods and expert's knowledge
  • Methods for variable selection and construction, and discovery of interactions
    • Common scenarios in data mining:
    • Rare event prediction
    • Working with unbalanced classes
    • Cost-sensitive prediction
    • Recommendation systems
    • Churn prediction and similar problems
  • Introduction to popular data mining tools

3. Readings

  • F. Witten, E. Frank: Data Mining: Practical Machine Learning Tools and Techniques, Morgan Kaufmann, 2005.
  • S. Few: Now You See It: Simple Visualization Techniques for Quantitative Analysis, Analytics Press, 2009.
  • E. R. Tufte: The Visual Display of Quantitative Information, 2nd Edition, Cheshire, CT: Graphics Press, 2001.

4. Intended learning outcomes

Knowledge and understanding:

  • Knowledge and understanding of data mining methods, ability to use them and evaluate the results.

Application:

  • Application on real-world data.

Reflection:

  • Understanding the relation between the theoretical aspects and practical use of the methods.

Transferable skills:

  • The course represents the pre-condition for the course in decision systems.

5. Learning and teaching methods

Lectures, exercises, homeworks and other assignments, practical work on artificial and real-world data.

6. Assessment

Type (examination, oral, coursework, project):

  • Continuing (homework, midterm exams, project work) (50 %)
  • Final (written and oral exam) (50 %)

Grading: 6-10 pass, 1-5 fail.