Lecture in Winter 2018/2019,
Paris Lodron University of Salzburg, Austria
On this web page you can find information about the lecture
"Data Mining / Intelligent Data Analysis"
that is given by Christian Borgelt
in winter 2018/2019 at the Paris Lodron University of Salzburg, Austria.
This page will be updated in the course of the semester.
| Type of activity: |
 |
Lecture/Exercise Lesson |
| Semester week hours: |
|
2 (total: 32 hours) |
| Credits: |
|
? (to be determined) |
| Language: |
|
German or English according to preference |
| Requirements: |
|
Standard university mathematics |
| Certificate: |
|
oral examination
(graded: ca.~30 minutes,
ungraded: ca.~10-15 minutes) |
Contents:
- Data and Knowledge, Knowledge Discovery in Databases,
Data Mining, Intelligent Data Analysis
- Descriptive Statistics: Tabular and Graphical Representations,
Characteristic Measures, Principal Component Analysis
- Inductive Statistics: Parameter Estimation, Hypothesis Testing,
Model Selection
- Regression: Linear, Multivariate Polynomial, and Logistic
- Decision and Regression Trees: Induction Algorithm, Attribute
Selection Measures, Pruning, Random Forests
- Artificial Neural Networks: Multilayer Perceptrons,
Function Approximation, Training Methods
- K-Nearest Neighbor: Lazy Learning, Number of Neighbors,
Attribute Weighting and Selection, Access Data Structures
- Clustering: k-Means, Learning Vector Quantization,
Fuzzy Clustering, Estimation of Mixtures of Gaussians,
Hierarchical Agglomerative Clustering
- Frequent Item Set Mining: Item Sets and Association Rules,
Mining Algorithms
Learning Objectives:
- Knowledge of the basic ideas of data mining, intelligent data
analysis and knowledge discovery in databases
- Knowledge of the most frequently used data mining methods
- Understanding of the advantages and disadvantages of different
data mining methods
- Ability to select a data mining methods for a given problem
- Ability to build models with data mining methods
- Application of data mining methods in practice
| Day of the Week |
Time |
Room |
Start |
|---|
| Monday |
15:00-17:00 |
HS 415 (H34OG1.A-002) |
01.10.2018 |
If you have questions about the lecture or the exercises,
please refer (by email or in person after a lecture) to:
Lecture slides (in English, version 2018.10.08, 464 slides).
| ida.pdf |
(2186 kb) |
one slide per page |
| ida4.pdf |
(2214 kb) |
four slides per page, two by two |
| ida4s.pdf |
(2213 kb) |
four slides per page, four by one |
| ida.zip |
(5221 kb) |
LaTeX and METAPOST source files |
Here exercise sheets with several exercises will be made available,
a selection of which will be discussed during the lecture.
The exercise sheets as well as their (future) solutions
are only available in German.
-
Guide to Intelligent Data Analysis:
How to Intelligently Make Sense of Real Data
Michael R. Berthold, Christian Borgelt, Frank Höppner,
and Frank Klawonn
Springer-Verlag, Berlin, Germany 2010, ISBN 978-1-84882-259-7
(397 pages, in English)
- Data Mining: Practical Machine Learning Tools and Techniques
Ian H.~Witten, Eibe Frank, Mark Hall, and Christopher Pal
Morgan Kaufmann, Burlington, CA, USA 2016, ISBN 978-0-12-804291-5
(654 pages, 4th edition, in English)
- Data Mining: Concepts and Techniques
Jiawei Han, Micheline Kamber, and Jian Pei
Morgan Kaufmann, Burlington, CA, USA 2011, ISBN 978-0-12-381479-1
(744 pages, 3rd edition, in English)
© 2018
Christian Borgelt
