Modulcode:	infADMMLM-01a
Englische Bezeichnung:	Advanced Data Mining and Machine Learning Methods
Modulverantwortliche(r):	Prof. Dr. Peer Kröger
Turnus:	jedes Jahr im SS (SS21 SS22 SS23 SS24)
Präsenzzeiten:	4V 2Ü
ECTS:	8
Workload:	60 Std. Vorlesung, 30 Std. Präsenzübung, 150 Std. Selbststudium
Dauer:	ein Semester
Modulkategorien:	MSc-Inf-WP (MSc Inf (21)) 2F-MEd-Inf-WP (MEd-Hdl Inf (21)) 2F-MA-Inf-WP (2F-MA Inf (21)) MSc-WInf-WP-Inf (MSc WInf (21)) WI (MSc Inf (15)) WI (MSc WInf (15))
Lehrsprache:	Englisch
Voraussetzungen:	Inf-ADS Inf-IS infKDDM-01a

Kurzfassung:

Basic data mining and machine learning algorithms are designed to work on tabular structured data objects also know as feature vectors. This module introduces and discusses advanced methods from data mining and machine learning for analyzing data which is not in the form of feature vectors. Examples for such data includes high-dimensional data, data streams, time series, image data, text data, graph data, etc.

Lernziele:

After completing the course, students should

• have a basic overview over techniques for analyzing non-vector data using data mining and machine learning methods including capabilities and limitations of the different approaches,
• be able to apply data mining and machine learning techniques new problems.

Lehrinhalte:

Selected content from the following general topics:

• High-dimensional data
  
  • Curse of dimensionality
  • Feature extraction, feature selection, dimensionality reduction
  • Unsupervised learning for high-dimensional data
• Mining sequence data
  
  • Similarity models for time series and sequence data
  • Learning in "real-time"
  • Data mining and machine learning for spatio-temporal data
• Analyzing graph-structured data
  
  • Similarity models for graphs
  • Graph Kernels
• Respresentation learning

Since the field of data mining and machine learning on complex data is currently a very volatile field with high innovation pace, this modul will cover current techniques and aspects of the listed topics.

Weitere Voraussetzungen:

Students should have profound knowledge from the following basic moduls:

• Informationssysteme
• Knowledge Discovery and Data Mining

Prüfungsleistung:

• Regular submission of and passing the laboratory/practical assignments
• Written exam

Lehr- und Lernmethoden:

Beamer presentation and use of software tools. 

Verwendbarkeit:

Literatur:

Up-to-date literature relevant to the course will be given in the lecture.

Verweise:

Kommentar: