From EContent to ELearning in Computational Linguistics - PowerPoint PPT Presentation

About This Presentation
Title:

From EContent to ELearning in Computational Linguistics

Description:

Development of Multi-Lingual Resources and Applications for Distant Learning. 11. Conclusion ... be done through the development of multi-lingual applications ... – PowerPoint PPT presentation

Number of Views:43
Avg rating:3.0/5.0
Slides: 12
Provided by: kiril
Category:

less

Transcript and Presenter's Notes

Title: From EContent to ELearning in Computational Linguistics


1
From E-Content to E-Learning in Computational
Linguistics
  • Localisation of Teaching materials for less
    processed languages

Kiril Simov, Petya Osenova,, Nikolai Vazov
BulTreeBank Project, Linguistic Modelling
Laboratory, Bulgarian Academy of Sciences and
Sofia University St. Kliment Ohridski
2
Plan of the Presentation
  • Our Research Expertise
  • Our Teaching Expertise
  • Combining Research and Teaching
  • E-learning for Less Processed Languages
  • Conclusion

3
Research Expertise
  • Language Resources for Bulgarian
  • Text Archive, Treebank, Morphological Lexicon,
    Gazetteers, Valency Lexicon, Semantic Lexicon
  • Language Processors for Bulgarian
  • Morphological Analyser, Disambiguator, Chunk
    Grammars
  • Language Processors for French
  • Time Expression Grammar

4
Teaching Expertise
  • CLaRK Programme
  • Tübingen-Sofia International Graduate Programme
    in Computational Linguistics and Represented
    Knowledge
  • Summer schools
  • Tübingen (1999), Sozopol (2000) and Sozopol
    (2002)
  • Electronic teaching materials
  • Lecture notes in Computational Morphology,
    Declarative Knowledge Representation and
    WordNets Principles and Applications
  • Project-oriented masters and PhD student
    education
  • POS disambiguation, XML Corpus Linguistics,
    Web-based materials

5
Combining Research and Teaching (1)
  • Bachelor Seminar on Computational Corpus
    Linguistics at the Department for Bulgarian
    Language, the Sofia University St. Kl. Ohridski
  • Formal languages, XML, Disambiguation, Chunk
    Grammars
  • Master Programme in Computational Linguistics at
    the French Department, the Sofia University St.
    Kl. Ohridski
  • A full programme covering Mathematical Methods,
    Logic, Linguistic Theories, Corpus Linguistics,
    Programming

6
Combining Research and Teaching (2)
  • The best results have been achieved by students
    who were also involved in a research project
  • In such an environment they receive
  • support in acquiring additional knowledge
  • practical experience in solving research problems
  • possibilities for further realisation in academic
    or industrial institutions

7
The main Problem of Teaching CL (in Bulgaria)
  • There is no Institution devoted to research and
    education in Computational Linguistics
  • The experts in the area work in different places
  • Solution
  • Multi-institutional research and teaching teams
  • Distant E-learning
  • International co-operation

8
E-learning for Less Processed Languages (1)
  • We consider the teaching in CL at two levels
  • Basic level mathematical methods in CL,
    linguistic models, basic technologies
  • Advanced level linguistic theories and treatment
    of specific language phenomena, machine learning,
    applications

9
E-learning for Less Processed Languages (2)
  • Basic level
  • Existing teaching materials (in other languages)
    can be used directly or after translation
  • Advanced level
  • Existing teaching materials need adaptation to
    Bulgarian language
  • We call the process of adaptation to Bulgarian
    Localisation of teaching materials

10
From E-Content to E-Learning
  • We see our role in a joint project for distant
    learning in CL in the following way
  • Localisation of existing teaching materials to
    Bulgarian
  • Providing Bulgarian Language Resources and
    Processing Modules
  • Development of Multi-Lingual Resources and
    Applications for Distant Learning

11
Conclusion
  • Distant learning in CL can cover the existing
    institutional and/or expertise gaps
  • The less-processed languages can provide
    (interesting or different) language data and
    processing modules
  • Localisation of teaching materials can be done
    through the development of multi-lingual
    applications
Write a Comment
User Comments (0)
About PowerShow.com