The course belongs to the MA Programme Linguistic Diversity in the Digital Age
- study track: language technology
- modules: Studies in Language Technology (LDA-T3100), Essentials in Language Technology (LDA-TA500), Comprehensive specialization in Language Technology (LDA-TB500)
This is an obligatory course for students of the study track in language technology, and elective for the students of LDA-TA500 and LDA-TB500.
The course is available to students from other study tracks and degree programmes.
Students are advised to take this course in year 1 (semester 2) or year 2 (semester 3). The course is offered once every year either during the spring term in period IV or during the autumn term in period I or II
After successfully completing the course, you will be able to
- explain common computational models of meaning representation
- describe the main principles of symbolic methods in connection with logic and compositional semantics
- describe the main principles of distributional semantics and representation learning in connection with vector space models
- discuss issues of non-compositionality and idiomatic expressions
- explain common methods for automatic word sense disambiguation, semantic role labeling, sentiment analysis and distributional representation learning.
- The course consists of lectures, lab sessions, tutorials and seminars.
- There may also be a final exam.
- Some sessions require student attendance.
- Semantics and pragmatics or equivalent (BA level)
- Programming for linguists or equivalent (BA level)
- Mathematics for linguists or equivalent (BA level)
- Machine learning for linguists or equivalent (BA level)
|Recommended optional studies
- Linguistics in the digital age (MA level)
- Computational Morphology (MA level)
- Computational Syntax (MA level)
The meaning of words and phrases can be modeled computationally from rather different perspectives. This course provides an overview of the field covering topics, such as:
- Meaning representations
- Symbolic, subsymbolic and analogical representations
- Lexical semantics, word senses, lexico-semantic relations (WordNet)
- Compositional semantics, semantics and logic, semantic roles
- Distributional semantics, vector-space models
- Non-compositionality, multiword/idiomatic expressions (rhetorical figures, metaphors, sentiment).
There is also hands-on experience with some relevant algorithms, for instance:
- Word-sense disambiguation
- Semantic role labeling
- Sentiment analysis
- Representation learning
- Word association and multiword expressions
|Study materials and literature
- Jurafsky, Daniel; Martin, James H.: Speech and language processing : an introduction to natural language processing, computational linguistics and speech recognition, Prentice Hall cop. 2009. 2nd ed (especially for symbolic representations and methods)
- Web material and material distributed on the course
|Activities and teaching methods in support of learning
- There are lectures, lab sessions and interactive seminars.
- The students work both individually and by exchanging ideas in larger groups.
- Exercises, instructions and additional course material are published on a web-based learning platform (Moodle).
|Assessment practices and criteria
Grading follows the standard scale 0 – 5.
The following aspects are taken into account in grading
- Performance in the exercises
- Performance in the seminar work
- Performance in the final exam
- Active participation during sessions.