Deep Generation
http://www.lt-world.org/hlt_survey/ltw-chapter4-3.pdf
- Natural Language Generation Group at the Department of Computing Science, University of Aberdeen (NLG Group Aberdeen)
- German Research Center for Artificial Intelligence (DFKI)
- Natural Language Technology Group (NLTG)
- Chris Mellish
- Helmut Horacek
- Donia Scott
- Mariët Theune
- Matthew Stone
- Robert Dale
- Kathleen R. McKeown
- Anja Belz
- David McDonald
- Tilman Becker
- Emiel Krahmer
- Claire Gardent
- John A. Bateman
- Michael Elhadad
- Richard Evans
- Johanna D. Moore
- Kees Van Deemter
- Cecile L. Paris
- Manfred Stede
- Michael White
- Ehud Reiter
- Charles Callaway
- KPML
- What You See Is What You Meant (WYSIWYM)
- Multimedia Abstract Generation for Intensive Care (MAGIC)
- A Reference Architecture for Generation Systems (RAGS)
- IDAS
- RealPro
- ProjectReporter
A knowledge-based approach to natural language generation that stresses theoretical motivation and re-usability of technology and knowledge sources across tasks and domains of dicourse. Opposed to shallow generation, which emphasizes rapid application development at the cost of genericity. Deep generation can be subdivided into principled approaches to
- Document planning - the subdivision of information onto clause-sized chunks and the semantic and rhetorical structuring for presenting that information;
- Sentence planning - the choice of words, aggregation of information into linguistic units, and the generation of referring expressions;
- Surface generation - the realization of an actual text based on these decisions, including linearization, addition of closed class words, and morphological inflection. See also Syntactic Generation.