RDFferret Search Technology for the Semantic Web

1
RDFferretSearch Technology for the Semantic Web

• Deliverable 11 - Search Facility
• Uwe Krohn
• BT

2
Ontoknowledge Architecture
RDFferret
3
RDFferret RDF Full Text

• Impractical to create RDF annotations that
  exhaustively cover the content of a given
  document
• RDF searches might produce low recall
• RDF searches produce high precision
• Search both RDF annotations and text content
• Use well tested IR techniques (stemming, ranking)

4
RDFferret Main Features

• Full text search and RDF querying
• Continuity of full-text retrieval with structured
  approach
• Ontology-based indexing
• Ontology-based querying
• Resource-based indexing finer granularity of
  information retrieval
• Simple navigation of ontology
• Low threshold - high ceiling

5
RDFferret WEBferret
RDFferret
WEBferret
Ontology-Based Indexing
Database Management System
Structured Index
6
Ontology-Based Indexing
Property
Literal
Last_name
Pablo
Style
Painter
Cubism
First_name
Picasso
Resource type
RDF resource
Descriptors
7
Ontology-Based Index
Multi-Dimensional Index
Descriptor 1
Resource A
Descriptor 2
Resource B
Descriptor 3
Resource C
Descriptor 4
Resource D
Descriptor 5
8
Ontology-Based IndexSchematic Example
Resource type
Property
Keywords
Pablo
Painter
text
Pablo
Painter
First_name
Picasso
Painter
text
www.art.nl
Picasso
Painter
Last_name
Painter
Painter
text
Cubism
Painter
Style
9
RDFferret Demo