Knowledge Yielding Ontologies
for Transition-based Organization
  • Increase font size
  • Default font size
  • Decrease font size

WordNets in KYOTO

Wordnet play a key role in the processing in KYOTO:

  1. They provide the mapping from text to concepts,
  2. They allow for the sharing of a single central ontology across the languages
  3. They allow for the semantic organization of the automatically learned terminology
  4. They allow us to connect background vocabularies to the KYOTO system

The following modules in KYOTO use data from wordnets:

  1. Multiword tagger reads a Wordnet-LMF file to find the wordnet multiwords in KAF
  2. UKB word-sense-disambiguation program uses a lexicon and graph of semantic relations that are derived from wordnet
  3. Tybot term extraction uses wordnet concepts to semantically organize terms
  4. Named-Entity-Recognition uses wordnet to determine the type of named-entity and to find generic locations such as references to rivers, valleys, coasts, etc.
  5. Onto-tagger inserts ontological implications into KAF on the basis the wordnet structure
  6. Kybot fact extraction uses profiles that can be defined using synsets from wordnets
  7. KYOTO semantic search uses the wordnet hierarchy to carry out a conceptual match between the query and the facts
  8. KYOTO semantic search uses the equivalence relation in the wordnets to generate translations of facts in cross-lingual search
The wordnets present interface between the domain vocabularies and the ontology in the 3-layered knowledge architecture.

ICT-211423 - 2008 © Kyoto Consortium