Giuseppe Pirrò - Graph Navigational Languages: Harnessing the Power of Structured Data on the Web

13:00
Tuesday
18
Feb
2014
Organized by: 

Jérôme Euzenat (EXMO)

Speaker: 

Giuseppe Pirrò

Teams: 

- Salle A103, Inria Grenoble Rhône-Alpes, Montbonnot

- Short Bio of Giuseppe Pirrò, The Free University of Bozen-Bolzano, Italy, Giuseppe.Pirro@unibz.it

Giuseppe Pirrò is an assistant professor at the Free University of Bozen-Bolzano (Italy) since May 2011. Before joining FUB he was a postodoc (ERCIM Alain Bensoussan fellow) at INRIA Grenoble, France. He obtained his PhD at the DEIS Department, University of Calabria, Italy in 2009. His research interests lie in the areas of the Semantic Web, Graph Query Languages, Artificial Intelligence and Distributed Systems. He has authored about 30 papers published at top conferences and journal including CIKM, WWW, ISWC, AAAI, CCGRID, FGCS, and DKE. He serves as program committee member of conferences and workshops in the area of the Semantic Web and is involved in local and international research projects.

The Web, traditionally viewed as a vast repository of documents, is being transformed into a huge database by the massive presence of structured data. There are several efforts to publish, describe, organise and interlink structured data on the Web. New research challenges arise in this information space due to the intrinsic decentralised data creation and management, the lack of superimposed schema, the unknown topology of data sources and their links, and the availability of huge volumes of data covering diverse domains. This unbounded and distributed graph of structured data is usually called the Web of Data. The broad objective of this talk is to show the need for formal languages and (automatic) tools to leverage the wealth of structured information available on the Web of Data graph.

I will introduce a navigational language, called NautiLOD, which enables the declarative specification of data sources (nodes) on the Web of Data graph as well as the possibility to command actions over data. NautiLOD has a simple syntax and a formal semantics that captures both nodes in the Web graph as well as regions (i.e., subgraphs). It is the ideal language for Web developers and users that want to access Web data sources. NautiLOD has been implemented in the swget tool, which is available for download at http://swget.wordpress.com.

I will then introduce a proposal that formalises the notion of region and map for the Web. As in traditional cartography maps emerged as a means to help orient users in large spaces, Web maps can help users to orient and harness the potentialities of the Web space. Via Web maps, users can track, record, identify and abstract conceptual regions of information on the Web, for their own use, for sharing/exchanging with other users and/or for further processing (e.g., combining with other maps). Web maps are also useful to analyse information and visualise relationships and patterns that might otherwise remain hidden in the massive amount of data available on the Web. The formal grounding of the proposal makes maps suitable to be constructed, interpreted and manipulated in a semi-automated way. A tool implementing the map proposal is available for download at http://mapsforweb.wordpress.com.