Towards the implementation of a refined data model for a Zulu machine-readable lexicon
Download as PPTX, PDF0 likes388 views
This document discusses refining the data model for a machine-readable Zulu lexicon. It proposes a lexicon update framework using a morphological analyzer and guesser to identify new stems and roots from corpora. A data model is developed representing verbal extensions, deverbatives, and their morphological structures. Native XML and object-oriented databases are considered suitable for implementing the data model due to their ability to represent semi-structured, recursive data and provide different views and sequential access to the morphological information. Future work includes developing and evaluating prototypes for the Zulu lexicon and software to semi-automate the lexicon update framework.
Towards the implementation of a refined data model for a Zulu machine-readable lexicon
- 1. Towards the implementation of a refined data model for a zulu machine-readable lexiconRonell van der MerweLaurette PretoriusSonja Bosch1
- 2. Goal ... to develop ....a complete data repository (database)Bantu languagesMachine Readable (MR) lexiconapplications: finite state morphological analyser, ....2
- 3. What we will be discussing ....Brief overviewLexicon update frameworkData modelImplementation approachesConclusion Future work3
- 4. 1. Introduction - Comprehensive MR lexiconUpdate frameworkZULU corporaPaper Dict4
- 5. 2. Lexicon Update FrameworkMorphological AnalyserZulMorphCorpusapplyfailuressuccessesrebuildGuesserMR lexiconupdatenew stems/roots5
- 6. Purpose of MR LexiconMorphologicalAnalyserAppsZulMorphMR lexiconusedRepository of all lexical informationre-usedApps6Morphosyntactic information:Zulu morphotacticsMorphophonological alternation rulesEmbedded stem/root lexicon
- 7. Guesser variantPhonologically possible stems /rootsGuesseridentifynew candidate(s)MR lexiconupdate / include7
- 8. 3. Data Model - development8* 0 or more+ 1 or more| optional leaf node
- 9. 3.1 Verbal extensionsSuffixing extensions to verb root:AppliedCausativeIntensiveNeuterPassiveReciprocal9* 0 or more+ 1 or more| optional leaf node
- 10. 10
- 11. 11
- 12. 12
- 13. 3.2 Deverbativesverb / extendedverb rootnoun class prefixdeverbative suffix (-o- | -i- |-a-| -e-| -u-)13optional:nominal suffix(Aug | Dim | Loc | Fem)
- 14. 14
- 15. 15
- 16. 4. Towards implementation16Morphological AnalyserZulMorphMR lexiconThe capturing of the data must be:rigoroussystematicappropriately structuredTo ensure that:data exchange is consistent
- 17. Considerations in the choice of database...type of datadifferent views of datadifferent types of access to data17
- 18. Consideration: Type of Datasemi-structuredallows for recursionn-depth structures 18
- 19. Consideration: views and accessRebuilding the Morphological Analyser:view: morphological structure of all word roots/stemsaccess: sequential19
- 20. XML – enabled database20Object-Oriented DatabaseRelational Database<root>bon</root><verbfeatures>.......</verbfeatures>
- 21. Native XML database (NXD)21Native XMLDBXQueryXUpdate<root>bon</root><verbfeatures>.......</verbfeatures>
- 23. NXD and OO Database satisfy the implementation considerations:Both ...are suitable for our type of data: semi-structured , recursivesatisfy the type of access and view: sequential access to morphological information of all word roots/stems 23
- 24. ConclusionRefinement of a data model for the Zulu MR lexiconverbal extensions and deverbativesMR Lexicon embedded in an update frameworkNative XML and OO Databases possible approaches to implementation24
- 25. Future workDevelopment and evaluation of prototypes for the MR Lexicon (Zulu)Software for semi-automating the lexicon update frameworkBootstrapping of MR Lexicon for other Bantu languages25