To get EDIT-rights for the wiki, please create an account and contact Alice Carpentier (alice.carpentier@sti2.at) with your name, affiliation and username to be approved!

Data Streams and dynamicity

From PlanetData - WIKI
Jump to: navigation, search

Title: Data Streams and dynamicity

Number: WP01

Activity: Activity A1 Research

Activity leader:
   
Activity A1 Research Grigoris Antoniou

Lead partner: CWI

Work package leader: ???

This page has been generated with no predefined structure by Oscar Corcho, so as to provide some descriptions of the work being carried out in the context of this WP.

W3C Semantic Sensor Network Ontology

The ontology is described and available from here and a Journal of Web Semantics paper describing it is available at [1]

Ontology-based access to data streams

We have continued working on the SPARQL-Stream system. Papers related to this work are available at [1] (presentation), [2]

[1] Jean Paul Calbimonte, Oscar Corcho, Alasdair J Gray. Enabling Ontology-based Access to Streaming Data Sources. Proceedings of 9th International Semantic Web Conference (ISWC 2010). November 2010. [2]

[2] Enabling Query Technologies for the Semantic Sensor Web. J.-P. Calbimonte, H. Jeung, O. Corcho and K. Aberer. International Journal on Semantic Web and Information Systems IJSWIS, Volume 8(1). IGI Global, 2012. [3]

Ontology-based characterisation of data streams

We have also done some work in this direction, as described in [3] and [4]

[3] Deriving Semantic Sensor Metadata from Raw Measurements. J.-P. Calbimonte, Z. Yan, H. Jeung, O. Corcho and K. Aberer. In Proc. of the 5th International Workshop on Semantic Sensor Networks SSN 2012, at ISWC 2012. Boston, USA, Nov 2012. [4]

[4] Semantic Sensor Data Search in a Large-scale Federated Sensor Network (Best Paper Award). J.-P. Calbimonte, H. Jeung, O. Corcho and K. Aberer. In Proc. of the 4th International Workshop on Semantic Sensor Networks SSN 2011, at ISWC 2011, Bonn, Germany, Oct 2011. PDF presentation

Benchmarking streaming query engines

The work done in this direction has been published in an ISWC2012 paper [5]

[5] SRBench: A Streaming RDF/SPARQL Benchmark. Y. Zhang, M.-D. Pham, O. Corcho and J.-P. Calbimonte. In Proc. of the 11th International Semantic Web Conference ISWC 2012. Boston, USA, Nov 2012 [5]

Next steps

From the second year review, we obtained the following feedback from the reviewers, which we will be addressing in year 3:

  • The 17 queries (BGP matching, solution modifier, query form, streaming, SPARQL 1.1, reasoning) defined in the SR benchmark are a good first step but it is unclear how they were motivated, hence we suggest to devise and follow a methodology for the query selection. The connection between the benchmarking work by CWI (LDBC project) and the use cases from UPM should be better connected.
  • The broader question about what is the impact of RDF streams should be addressed. We suggest defining how to 'sell' to industry, i.e. providing arguments as to why this sort of stream systems are needed (cf. in-memory solutions like SAP Hana).
  • Try to collaborate/coordinate efforts with the Smart Vortex project and the work on semantic continuous queries performed in Uppsala (by Tore Risch). This would include a more principled approach for involving external partners.
  • Could existing streaming systems like S4/Storm/Kafka be used or leveraged? If not, why not?