Language Processing for Different Domains and Genres (WS 2009/10)

Schedule (Note this schedule is likely to change, esp. wrt the later presentation topics)

Date	Topic / Speaker	Comments / Slides
15.10.09	Introduction	Intro (.pdf) Intro to Linux (.pdf)
22.10.09	Presentation of taggers and parsers; exercises Note: change of room: CIP Pool Stanford POS Tagger (Kang) /proj/contrib/stanford-postagger-full-2008-09-28 TreeTagger (Lisa) /proj/contrib/workingTreetagger C&C tagger (Carolyn) /proj/contrib/candc-1.00 Stanford parser (lexicalised: Todd; unlexicalised: Martin) /proj/contrib/stanford-parser-2008-10-26 Berkeley parser (Daniel) /proj/contrib/berkeleyParser Some datasets for testing can be found here: /proj/contrib/lpdd/	Class Documentation TreeTagger C & C Tagger Stanford Tagger Berkeley Parser All three taggers use the PennTreebank tagset. See also the Annotation Guidelines. An introduction to Linux can be found here (by Machtelt Garrels).
29.10.09	continuation of presentations; exercises	Class Documentation Stanford parser (lexicalised) Stanford parser
05.11.09	presentation of WSD tool	MaJo - WSD Tool (slides) MaJo - Documentation NLP tools (overview) PCFGs and Parsing (overview)
12.11.09	machine learning tutorial; visualisation	Introduction to Machine Learning Data Visualisation with R
19.11.09	Domain/Genre Differences (1) general: Biber 1993 (Martin)	Biber: Register Analysis
26.11.09	Note: Konferenzraum 2.11! Domain Adaptation for ML discourse: Webber (Todd) Daumé 2007 (Kang)	Presentation: Webber 2009 Presentation: Daumé 2007: Domain Adaptation
03.12.09	Parsing (1) Reranking: McClosky et al. 2006 & Foster et al. 2007 (Maria) Self Training: McClosky et al. 2008(Marcel)	Presentation: Reranking (McClosky et al. 2006, Foster et al. 2007) Presentation: Self-Training (McClosky et al. 2008)
10.12.09	Parsing (2) detection of non-generalising rules: Dickinson & Foster 2007 (and possibly Dickinson 2008) (Fai) Lexicalised parsing for different domains. Laura Rimell, Stephen Clark. 2008.(Daniel)	Parsing Presentations: Detecting non-generalising rules Presentation: Lexicalised Parsing Presentations of Projects: POS Tagger Adaptation Parsing
17.12.10	Word Sense Disambiguation Most Frequent Sense: Koeling et al. 2005 (Lisa) Active Learning: Chan & Ng 2007 (Jonas)	WSD presentations: Most Frequent Sense
07.01.10	Project work
14.01.10	project work
21.01.10	project work
Course Registration Deadline: 22.01.2010!
28.01.10	project work
04.02.10	project work	Exam week?