Computational Lexicons and the Semantic Web Training Session

1
Computational Lexicons and the Semantic
WebTraining Session

• Alessandro Lenci
• Università di Pisa Department of Linguistics
• Istituto di Linguistica Computazionale - CNR

2
Case Studies

• Formalize in RDF the EuroWordNet Synsets and Top
  Ontology
• Writing semantic frames in RDF/S as basis for
  interlingua representations

3
EuroWordNetTop Ontology

• Top Concepts (TC) classify synsets
• the subset of 1024 Basic Concepts synsets
• TC are hierarchically ordered by means of a
  subsumption relation
• multiple inheritance between TC is not allowed
• BC can be cross-classified in terms of multiple TC

It is important to realize that the TC are more
like semantic features than common conceptual
classes Vossen (2001)
4
EuroWordNet
5
EuroWordNetTop Ontology
6
Task 1

• Formalize in RDF/S the EWN TC
• Creating RDF/S entries for the following synsets
  and link them to the relevant TC
• fruit 0001
• TC Comestible ? Function Object?Form
  Part?Composition Plant?Living?Natural?Origin
• cell 0002
• TC Part?Composition Living?Natural?Origin
• car 0003
• TC Artifact?Origin Object?Form Vehicle?Function

7
Lexical ObjectsTop Level
mlmLexObject
rdfssubClassOf
rdfssubClassOf
mlmSynset
rdfssubClassOf
mlmEntry
mlmhasSynU
mlmcorrespondsToSynset
mlmhasSemU
mlmSynU
mlmSemU
mlmlanguage
rdfsliteral
8
Task 2

• Formalize in RDF/S the following type of semantic
  frame
• share semantic frames among entries in different
  languages

9
Task 2semantic frames
PREDemploy1 Arg1ltAGENT - HUMANgt Arg2ltPATIENT
- HUMANgt
agent nominalization
master link
patient nominalization
event nominalization
SemU employee
SemU employment
SemU to employ
SemU employer