DanTermBank 
Phase 1: Establishing a Danish Terminology and Knowledge Base

The aim of the project is to develop methods for automatic knowledge extraction, automatic construction and updating of ontologies. In the project methods will be developed for automatic merging of terminological data from various existing sources, as well as methods for target group oriented knowledge dissemination.  The research carried out in the current project is a prerequisite for establishing a national Danish term bank which can ensure development and quality of Danish LSP. When the term bank has been established, it will form the basis for various other research projects.

This phase of the project is supported by  VELUX-fonden

1. Knowledge Extraction.
The aim of this subproject is to develop new advanced models of and methods for automatic extraction of concepts and information about conceps as well as a prototype which can automatically produce a draft version of a terminological ontology on the basis of an existing domain-specific text corpus, or on the basis of domain texts automatically collected from the Internet. Thus, the draft ontologies will contain subdivision criteria and characteristics as formal feature specifications on concepts

2. Knowledge Structuring.
The aim is to develop methods and a prototype that may be used for automatic validation and dynamic expansion of the draft ontologies that result from the automatic knowledge extraction.

3. Knowledge Dissemination.
This subproject will analyse possible user groups (e.g. experts, semi-experts and various groups of laymen) in order to target the structure and contents towards the communication-oriented and knowledge-oriented needs of different target groups.

Learn more about the project in the attached document (In Danish) DanTermBank wwwcbs and by visiting the project website: http://dantermbank.cbs.dk/