|
|
Spezialvorlesung "Machine Learning and Data Mining"
Prof. Dr. Luc De Raedt
Co-organizers: Andreas Karwath and Albrecht Zimmermann
|
Lectures
|
|
Wednesday 9-11 o'clock
|
|
Room: SR 00-014, Building 101
|
|
|
|
Friday 9-10 o'clock
|
|
Room: SR 00-014, Building 101
|
|
Exercises
|
|
Friday 10-11 o'clock
|
|
Room: SR 00-014, Building 101
|
Credit points (Kreditpunkte): 6
- The re-examination will be done orally and will last 20 minutes (only for those who have failed the written exam). To make the dates and times as flexible as possible for you, we would like every student to arrange an appointment with us via email. The examination will have to be this semester. Please suggest a time corridor and probably more than one day so we can fit you in. Start making an appointment by sending us an email to Andreas.
- If you want to get a certificate of attendence (a 'schein'), please get in contact with Susanne. The certificates are supplied uppon request.
The ability to learn is central to intelligence. Therefore machine learning
is one of the key areas in artificial intelligence. Machine learning is
concerned with building agents that can learn in one way or another. Various
techniques for learning will be surveyed in this course. This includes
:
-
symbolic learning
-
neural networks
-
genetic algorithms
-
decision trees
-
bayesian networks
-
versionspaces
-
inductive logic programming
-
computational learning theory
The most popular applications of machine learning are in the analysis of
data. In recent years, this has been the key topic of data mining. This
is a new science, which aims at discovering (nuggets of) knowledge from
data. To realize this, it combines machine learning, statistics and database
techniques. Special emphasis is laid on analysing large datasets. Techniques
from data mining and their applications will also be surveyed in this course.
This includes :
-
the knowledge discovery process
-
association rule discovery
-
handling large data sets
-
...
This course is especially relevant to BioInformatics as the analysis
and interpretation of
(large amounts of) data is one of the key problems in BioInformatics.
Furthermore, techniques from the field of machine learning and data mining
complement the more traditional data analysis techniques from statistics.
Various applications in BioInformatics will be discussed in the course.
Throughout the course the book by Tom Mitchell, Machine Learning,
McGraw Hill, 1997 will be used and complemented by other materials.
|
