Ruprecht-Karls-Universität Heidelberg
Institut für Computerlinguistik

Bilder vom Neuenheimer Feld, Heidelberg und der Universität Heidelberg

Topic Models

Kursbeschreibung

Studiengang Modulkürzel Leistungs-
bewertung
BA-2010 AS-CL 8 LP
NBA AS-CL 8 LP
Master SS-CL, SS-TAC 8 LP
Magister - -
Dozenten/-innen Vivi Nastase
Veranstaltungsart Hauptseminar
Erster Termin 26.04.2012
Zeit und Ort Do, 14:1515:45, INF 325 / SR 24 (SR)

Leistungsnachweis

  • implement a seminar project
  • pass a written exam

Inhalt

In this seminar we will understand what topic models are, and how they are useful for the processing of texts. We will study first the basic topic model (LDA), and then extensions of this in various dimensions.

Course organization:
In the first part of the semester I will give lectures, in the second part, the students will present and discuss papers that I will assign to them on the topic. Throughout the semester the students will have to implement a topic model, and give a demo and short presentation at the end of the course.

Kursübersicht

Seminarplan

Datum Sitzung Materialien
26.04.2012 No lecture (Vivi is away)
3.05.2012 Introduction to the course, probabilities refresher Step 1 of our course-long project due May 17th
10.05.2012 Building up towards LDA (partial slides) Reminder: Step 1 due May 17th
17.05.2012 Holiday (but the assignments are still due today!) Step 2 of our course-long project due May 31st
24.05.2012 LDA Step 3 of our course-long project due June 14th
31.05.2012 Student presentations 1. (Annika Berger) Reading tea leaves: how humans interpret topic models Chang, Boyd-Graber, Gerrish, Wang & Blei, 2009 (slides)
7.06.2012 Holiday Step 4 of our course-long project due July 12th
14.06.2012 Student presentations 2. (Schigehiko Schamoni) Probabilistic author-topic models for information discovery Steyvers, Smyth, Rosen-Zvi & Griffiths, 2004 (slides)
3. (Hans-Martin Ramsl) Studying the history of ideas using topic models Hall, Jurafsky & Manning, 2008 (slides)
21.06.2012 Student presentations 4. Supervised topic models Blei & McAuliffe, 2007
5. (Sariya Karimova) Topics over time: a non-Markov continuous time model of topical trends Wang & McCallum, 2006
28.06.2012 Student presentations 6. A Latent Dirichlet Allocation methods for selectional preferences Ritter, Mausam & Etzioni, 2010
7. (Anja Summa) Syntactic topic models Boyd-Graber & Blei, 2009
5.07.2012 No lecture (Vivi is away) Step 5 (optional) of our course-long project due at the end of the course
12.07.2012 Student presentations 8. (Angela Schneider) Extracting multi-lingual topics from unaligned comparable corpora Jagarlamudi & Daume, 2010
9. (Eleftherios Matios) A topic model for word sense disambiguationBoyd-Graber, Blei & Zhu, 2007
19.07.2012 10. (Benjamin Heinzerling) PCFGs, topic models, adaptor grammars and learning topical collocations and the structure of proper names Johnson, 2010
11. Continuous dynamic topic models Wang, Blei & Heckerman, 2008
26.07.2012 Student project presentation and discussion

Literatur

» weitere Kursmaterialien

zum Seitenanfang