Natural Language Processing
|(1)||25/8 13-15||Introduction. (MD, JN, MK)||9-2042|
|(2)||10/9 10.00-12||Information retrieval (JT). Slides.||ULL|
|(3)||8/10 10-12||Dialoge systems (JG)||ULL|
|(4)||22/10 10-12||Machine Translation (JT). Slides.||ULL|
|(5)||10/11 10-12||Words (RT) Slides.||ULL|
|(6)||11/11 10-12||Words (RT)||Chomsky|
|(7)||11/11 13-15||Words (RT)||Chomsky|
|(8)||16/11 10-12||Syntax (MK)||ULL|
|(9)||18/11 13-15||Syntax (MK)||Turing|
|(10)||23/11 10-12||Syntax (MK)||ULL|
|(11)||1/12 10-12||Semantics (MS) Slides||ULL|
|(12)||7/12 10-12||Semantics (MD) Slides||ULL|
|(13)||9/12 10-12||Semantics (MD)||Chomsky|
|(14)||13/01 10-12||Course round-up||9-2029|
ULL: Interurban Lectures given through Sunet's Adobe Connect server. Connect through:
Teachers: Mats Dahllöf, Joakim Gustafson, Marco Kuhlmann, Joakim Nivre, Magnus Sahlgren, Jörg Tiedemann, Reut Tsarfaty.
Themes and readings.
- November 10.
- Corpus-based linguistics and Levels of analysis.
- Regular Expressions and Finite State Morphology. J&M, ch 2-3.
- N-Grams and Language Modeling. J&M, ch 4.
- Word Classes and Part of Speech Tagging. J&M, ch 5.
- Extra: Joint morphological and syntactic analysis. Habash & Rambow, Tsarfaty.
- November 16-23. J&M, ch 12-14.2.
- December 1. Turney and P. Pantel
- December 7. J&M, ch 17-21.
Intended learning outcomes
In order to pass the course the student is required to be able to:
(1) give an account of the products, services and technologies that are typical for the language technology field, and provide details of their performance and commercial importance and give examples of common metrics to evaluate them;
(2) describe and implement methods for morphological analysis and tagging of natural language, and evaluate such systems;
(3) write formal grammars for a language-technological purpose;
(4) describe and implement some important parsing algorithms, and evaluate parsing systems;
(5) describe and implement some methods for capturing and/or classifying the content of texts in natural language.
Examination and grading criteria
The course outcome is examined by means of three (packages of) assignments, one each for the themes of Words, Syntax and Semantics. In order to pass the course you'll have pass each of one of these. In order to pass the course with distinction (Väl godkänt), you'll have to pass two of the three assignments with distinction. The requirements for each assignment are given in the assignment PM. The project-size assignment of the 5LN702 NLP Implementation Project will be counted as fullfillment of either one of the Words or Syntax assignments, given that the project has the corresponding emphasis.
- "Words" (Reut Tsarfaty): assignment, data files.
- "Syntax" (Marco Kuhlmann): assignment.
- "Semantics" (Mats Dahllöf): assignment.
Deadline: January 21, 2011.
Daniel Jurafsky and James H. Martin (2009), Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition. Second Edition, Pearson Education.
Habash, Nizar and Rambow, Owen (2005) Arabic Tokenization, Part-of-Speech Tagging and Morphological Disambiguation in One Fell Swoop. Proceedings of the 43rd Meeting of the Association for Computational Linguistics (ACL'05).
Reut Tsarfaty (2005), Integrated morphological and syntactic disambiguation for Modern Hebrew, Proceeding of SRW COLING-ACL.
P. D. Turney and P. Pantel (2010) From Frequency to Meaning: Vector Space Models of Semantics, Journal of Artificial Intelligence Research (JAIR), Volume 37, pages 141-188.