13.02.2013 Semantic Cluster

From IMarine Wiki

Jump to: navigation, search



Time: Monday February 13, 2013, 10:00 - 12:30 Europe/Rome

  • FLOD initiative;
  • TLO;


CNR: Leonardo Candela, Pasquale Pagano;

FAO: Claudio Baldassarre, Anton Ellenbroek, Fabio Simeoni;

Forth: Carlo Allocca, Pavlos Fafalios, Ioannis Kitsos, Yannis Marketatis, Yannis Tzitzikas;

IRD: Julien Barde;

Main Discussions and Actions

On collaboration / cooperation

  • Yannis highlighted the need to reinforce the exploitation of TRAC tickets (this is a project policy)
    • all agreed on reinforcing this;

On FLOD Initiative

  • Claudio presented shortly the intent of the FLOD initiative as the stream of activities with the objective to produce software tools, integrated in the infrastructure, that enables the users (case study is WoRMS and VLIZ) to produce Linked Data version of their dataset. Multiple converted dataset are interlinked to form the iMarine LOD of EA. One of the benefit is the distributed maintenance, aside the production of network of Linked Datasets. More details are in the slides.
  • slides: https://portal.i-marine.d4science.org/group/data-e-infrastructure-gateway/workspace?itemid=ee3f415e-681b-4973-aba5-c0a7770d8491
  • Claudio and Fabio are working on a document better describing the proposal and its integration with the rest of the iMarine infrastructure.

On potential clients / exploitation


  • what is the relation between FLOD and the TLO?
    • (Claudio) It seems that the proposed approach is a Top Level Knowledge Base rather than a Top Level Ontology;
    • (Yannis) the TLO is a structure/schema to might be used to query different KBs;
      • however, an evaluation step is needed to verify whether the TLO is powerful enough to represent/describe the content of the KBs;
      • we have to follow both the "materialized integration approach" and the "virtual integration approach";
    • (Claudio) the FLOD proposal discussed above is to provide the infrastructure with a toolkit for producing RDF by the users that choose the TLO as reference ontological model.
      • This can be used similarly to SPD, i.e. these RDF triples can be used to populate TLO;
      • (Yannis) In doind this, you should identify and use standards and namespaces. The TLO initiative is intended to promote this, i.e. to develop a shared / agreed "namespace" / "schema";
    • (Lino) One of the problems that I see with "materialisation" is the fact that data evolves and the materialisation should take this into account;

Agreed Actions Summary

  • FAO (Baldassarre/Simeoni) will circulate a document on FLOD initiative that specifies the requirements and assessment of the resources required, based on the case study of WoRMS and VLIZ dataset;
    • WP10 will analyse this proposal and allocate resources leading to the development of the technology described here;
  • On TLO development, it is a matter of involving the "community" in this modeling activity (FAO, IRD, etc.);
    • This will be discussed in the next SB;
Personal tools