TU BRAUNSCHWEIG
| Carl-Friedrich-Gauß-Fakultät | Informatik
Informatikzentrum

Machine Learning for Pervasive Computing

Modulnr.INF-KM-32
Veranst.Nr.INF-KM-43
Studieng.Master Informatik, Master Informations-Systemtechnik, Master Wirtschaftsinformatik
IBR Gruppe(n)CM (Prof. Wolf)
ArtVorlesung/Übung
Dozent
PhotoDr. Stephan Sigg
Researcher DAAD
sigg[[at]]ibr.cs.tu-bs.de
+49 531 3913249
LP5
SWS2+1
Ort & Zeit Lecture: Wednesday, 11:30-13:00 Raum IZ-358
Exercise: Wednesday, 13:15-14:45, Raum IZ-358
This lecture is discontinued. If you are interested in Machine Learning and Applications please consider the Lecture Machine Learning for Computer Security (Institute of System Security).
BeginnMittwoch 15. April, 11:30-13:00
Hörerkreis Studenten der Informatik und Elektrotechnik, Wirtschaftsinformatik, Informations-Systemtechnik, Medienwissenschaften, Studenten mit Nebenfach Informatik
Voraussetzungenkeine
Scheinerwerb Successful oral examination. Furthermore, active participation in the exercises is expected.
Inhalt

Lecture slides

Diese Unterlagen sind nur für registrierte Teilnehmer zugänglich. Um sich zu registrieren, benötigen Sie einen Account: entweder einen IBR-Account, den Sie vielleicht im Rahmen einer Arbeit am Institut bereits erhalten haben, oder einen selbst aktivierten IBR-y-Account. Anschließend melden Sie sich über die Login-Funktion (oben auf dieser Seite) an.
[ Podcast | Podcast aller Formate | iPod-Video-Podcast | Audio-Podcast | Newsfeed aller Formate ]
KapitelFolienFilmeAudioÜbungen
0. Organisation
pdfpdfpdf
1. Introduction
pdfpdfpdf
2. Rule based
pdfpdfpdf
3. Decision trees
pdfpdfpdf
4. Regression
pdfpdfpdf
exercise
5. Random search
pdfpdfpdf
6. High dimensional data
pdfpdfpdf
exercise
7. Artificial Neural Networks
pdfpdfpdf
exercise
8. Instance-based learning
pdfpdfpdf
exercise
9. Probabilistic graphical models
pdfpdfpdf
10. Topic models
pdfpdfpdf
11. Unsupervised learning
pdfpdfpdf
12. Anomaly detection, recommendersystems, online learning
pdfpdfpdf
Exkurs: Trendmining (Dr. Olga Streibel)
pdfpdfpdf
Exkurs: Clustering and density-based clustering (Nguyen Thach)
pdfpdfpdf

Lecture

We investigate popular algorithms applied for supervised and unsupervised machine learning on real data and consider test-set design principles, model and feature selection as well as performance metrics. The discussed considered cover linear/logistic/multivariate/multivariable regression, Support vector machines, artificial neural networks, k-nearest neighbour, k-means, self organizing maps, decision trees, naive Bayes and Bayesian networks, hidden markov models, conditional random fields and principle component analysis.

The course will address selected topics in Pervasive Computing with a special focus on Machine learning and activity recognition from sensor-data. In addition, other fields of Pervasive Computing are covered to provide students with a good overview on current advances and research challenges. Depending on the interest of the students, the emphasis on these additional topics may differ.

Course topics

  • Introduction to Machine learning
  • Supervised and Unsupervised learning
  • Features and feature extraction
  • Feature subset selection
  • Performance metrics
  • Polynomial curve fitting
  • Support Vector Machines
  • Artificial Neural Network learning
  • Clustering (k-means)
  • Dimensionality reduction
  • Anomaly detection
  • Recommender systems

Literature

  • Bishop, Christopher M. Pattern recognition and machine learning. Vol. 1. New York: springer, 2006.
  • Duda, Richard O., Peter E. Hart, and David G. Stork. Pattern classification. John Wiley and Sons, 2012.
  • Witten, Ian H., and Eibe Frank. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann, 2005.

aktualisiert am 05.04.2016, 08:28 von Dr. Stephan Sigg
printemailtop