NCSR at INDIGO

1
NCSR at INDIGO

• Vangelis Karkaletsis
• Kick-off Project Meeting
• Athens, 15 February 2007

2
A few words about NCSR

• National Centre for Scientific Research (NCSR)
  Demokritos
• Largest self-governing research organisation
  under the supervision of the Greek government
• Subdivided in 8 research institutes
• Biology, Material Sciences, Microelectronics,
  Informatics Telecommunications, Nuclear
  Technology Radiation Protection, Nuclear
  Physics, Radioisotopes Radiodiagnostic
  products, Physical Chemistry
• In INDIGO with the Institute of Informatics
  Telecommunications
• Software Knowledge Engineering Lab (SKEL)

3
and about SKEL
Information Integration
User-friendly information access
Ontology Creation and Maintenance
SKEL researchers aim to develop knowledge
technologies that will enable the efficient,
cost-effective and user-adaptive management and
presentation of information
4
Applied Research at SKEL

• The general-purpose language engineering platform
  Ellogon
• Language processing tools and resources
• The i-DIP platform for developing web content
  collection and extraction systems
• The QUAPRO proxy server, for validating RDF
  labels of web resources
• The FILTRON e-mail filter, that blocks
  unsolicited commercial e-mail (spam messages)
• The FilterX Web proxy filter, that blocks obscene
  Web content
• Tools for creating and maintaining ontologies
• The PServer general-purpose server for
  personalization
• Authoring tool for natural language generation
  systems

5
Basic Research at SKEL

• Ontology learning
• Ontology coordination
• Document Summarization
• Ontology-based information extraction
• Reasoning
• Natural language generation
• Web usage mining

6
SKEL team at INDIGO

• Vangelis Karkaletsis
• Senior researcher, Head of SKEL
• Georgios Paliouras
• Researcher
• Stasinos Konstantopoulos
• Post-doctoral researcher
• Thanasis Tegos
• MSc student
• Dimitris Bilidas
• Research associate

7
SKEL infrastructure for INDIGO

• Personalisation Server (PServer)
• General personalisation framework
• Independent of particular applications.
• Able to adapt to application parameters.
• Covers a wide range of services and
  functionality.
• Easily expandable.
• Easy to set up and use by ongoing projects.
• PServer operation
• Application initializes the service by defining
  organizing the particular features.
• Continually notifies PServer on user choices.
• Application can pose questions at any time

8
SKEL infrastructure for INDIGO

• Authoring tool for NLG applications
• Originally developed in the context of the M-PIRO
  project
• Being extended in cooperation with AUEB team
• Currently is able to
• Import OWL-DL ontologies
• Perform consistency checking using the reasoning
  engine RACER
• Export ontologies in OWL and linguistic and user
  modeling resources in RDF
• Interact with AUEBs NaturalOWL NLG engine (a
  previous version with UEDINs Exprimo NLG engine)

9
Extensions needed for INDIGO

• PServer
• Maintain both the users models and the robotic
  personality models
• Integrate new services
• Authoring Tool
• Interact with both NLG engines
• Support for spatial expressions
• Support for INDIGOs mark up scheme