From NLG Systems Wiki

Jump to: navigation, search

[edit] Summary


WYSIWYM aims to allow domain experts to encode their knowledge directly, by interacting with a feedback text, generated by the system, which presents the knowledge defined so far and the options for extending or revising it. Previous knowledge editors have provided a graphical interface so that users can interact with diagrams rather than writing code; WYSIWYM takes a step further, exploiting automatic text generation so that the user interacts with an ordinary natural language document rather than a relatively unfamiliar diagram.

The acronym means `What You See Is What You Meant'. The feedback text presented to the user (What You See) reveals the knowledge that has been encoded during the interaction so far (What You Meant). Documentation of knowledge bases becomes automatic, since the system is designed to produce a description in natural language of any knowledge base in any state of completion. The only limitation is that the knowledge base must conform to an ontology which from the user's point of view is fixed. If this ontology proves insufficient, it must be extended by a programmer; the user cannot add new concepts because the system would lack the linguistic resources to express them.


  1. Power, R., & Scott, D. (1998). Multilingual authoring using feedback texts. Paper presented at Proceedings of the 17th International Conference on Computational Linguistics and 36th Annual Meeting of the Association for Computational Linguistics, Montreal, Canada. Bib
  2. Power, R. (1999). Controlling logical scope in text generation. Paper presented at Proceedings of the 7th. European Workshop on Natural Language Generation (EWNLG'99), Toulouse. Bib
  3. Power, R. (1999). Generating referring expressions with a unification grammar. Paper presented at Proceedings of the 9th Conference of the European Chapter of the Association for Computational Linguistics (EACL'99), Bergen, Norway. Bib
  4. Scott, D. (1999). The Multilingual Generation Game: authoring fluent texts in unfamiliar languages. Paper presented at Proceedings of the 16th International Joint Conference on Artificial Intelligence (IJCAI'99), Stockholm, Sweden. Bib
  5. Piwek, P., Evans, R., & Power, R. (1999). Editing speech acts: a practical approach to human-machine dialogue. Paper presented at Preproceedings of Amstelogue'99: Workshop on the Semantics and Pragmatics of Dialogue, Amsterdam. Bib
  6. Kibble, R., Power, R., & van Deemter, K. (1999). Editing logically complex discourse meanings. Paper presented at Proceedings of the 3rd International Workshop on Computational Semantics, ITK, Tilburg University. Bib
  7. Power, R., Scott, D., & Evans, R. 1998. What you see is what you meant: direct knowledge editing with natural language feedback. Bib
  8. Scott, D. R., Power, R., & Evans, R. (1998). Generation as a Solution to its own Problem. Paper presented at Proceedings of the Ninth International Workshop on Natural Language Generation, Niagara-on-the-Lake, Ontario. Bib
  9. van Deemter, K., & Power, R. (1998). Coreference in knowledge editing. Paper presented at Proceedings of the COLING-ACL '98 Workshop on the Computational Treatment of Nominals, Montreal, Canada. Bib

Facts about WYSIWYMRDF feed
DescriptionWhat you see is what you meant': knowledge-editing with natural language generation  +
Domainknowledge editing  +
LanguageEnglish  +
Started1998  +
URLhttp://www.itri.bton.ac.uk/projects/WYSIWYM/wysiwym.html  +
WorkerPower  +, and Evans  +
Personal tools