DBCells - an open and global multi-scale linked cells

212 views

Published on

The land change models require large amounts of data, are difficult to be reproduced, as well as to be reused. Some initiatives to open and link data increase the reproducibility of scientific experiments and data reuse. One pillar of the linked data concept is the use of Uniform Resource Identifier (URI). In this paper, we propose DBCells – an architecture for publication of a global cellular space where each cell has a URI. This new approach will allow comparison, reproduction and the reuse of models and data. However, in order to succeed, this proposal requires participation, partnerships and investments. Our main purpose in this paper is to present the architecture, benefits and challenges for debating with the scientific community.

Published in: Science
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
212
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

DBCells - an open and global multi-scale linked cells

  1. 1. DBCells – an open and global multi-scale linked cells Universidade Federal do Maranhão Instituto Federal do Maranhão
  2. 2. Outline ● ● ● ● ●
  3. 3. Motivation
  4. 4. How to reuse model’s data ?
  5. 5. How to reuse model’s data ? Area expansion
  6. 6. Open Linked data ● ● ● Figure 1. From data to open linked data.
  7. 7. Open Linked data Figure 2 - Five-stars rating. Source: http://5stardata.info/en/
  8. 8. A global open linked data Figure 3 - Linked open data cloud. Source: lod-cloud.net. Last updated: 2014-08-30
  9. 9. Open Science
  10. 10. Dbcells dataset Publication layer Architecture: dbcells dataset
  11. 11. Cellular space
  12. 12. Cellular space Figure 4. TerraME and cellular space. Source: http://www.terrame.org/doku.php
  13. 13. Cellular space in the TerraLib
  14. 14. Cellular space
  15. 15. Divides the space in regular cells
  16. 16. Associates each cell to an identifier dbcells.org/cl015378
  17. 17. Represents each cell as an RDF graph dbcells.org/cl015378 dbpedia:resource/Altamira <dbcells:parent> -3.202778 -52.205833 100km2 <wgs84_pos:lat> <wgs84_pos:lon> <dbcells:resolution> DbCells Dataset DbPedia Dataset dbcells.org/cl015378 “The vision for RDF was to provide a minimalist knowledge representation for the Web”. (Shadbot et. al. 2006)
  18. 18. Dbcells dataset Publication layer Architecture: multi-scale relations Multi-scale relations
  19. 19. Multi-scale relations
  20. 20. Figure 5. Representation of strategies for spatial coupling in the case of regular cells. Source: [Moreira et al. 2008] Multi-scale relations
  21. 21. URI:http://dbcells.org/onto/relatedTo URI:http://dbcells.org/cl015378 URI:http://dbcells.org/cl6239 URI:http://dbcells.org/cl05356 URI:http://dbcells.org/cl5682 URI:http://dbcells.org/cl3278 Multi-scale relations
  22. 22. Models at different scales Climate Dataset URI:http://dbcells.org/cl015378 URI:http://dbcells.org/cl04379 Land Use Dataset Social Behaviour Dataset The relationships between cellular spaces allow the reuse of models at different scales and resolutions. For example, a land use model at a finer scale can use results of a climate model in a coarser scale
  23. 23. Dbcells dataset Publication layer Architecture: network relations Multi-scale relations Network relations
  24. 24. Network relations Figure 6. Schematic representation of a network-based spatial relation between cell objects in Scale 1 and point objects in Scale 2. Source: [Moreira et al. 2008]
  25. 25. Dbcells dataset Publication layer Architecture: models Multi-scale relations Network relations Land Use data set Climate data set The publication layer includes a dataset of cellular space, multi- scale relationships and open linked data models.
  26. 26. M1:cell1 dbcells:cl0151378 rdfs:sameAs PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> PREFIX owl: <http://www.w3.org/2002/07/owl#> PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> PREFIX xsd: <http://www.w3.org/2001/XMLSchema#> PREFIX dbcells: <http://dbcells.org/> PREFIX wgs84_pos: <http://www.w3.org/2003/01/geo/wgs84_pos#> SELECT ?y ?k WHERE {?x dbcells:parent ?y. ?k owl:sameAs ?x. FILTER regex(?y, "ATALAIA DO NORTE") } k y cells1 ATALAIA DO NORTE Architecture: models (linking to dbcells) sparql
  27. 27. Dbcells dataset Publication layer Architecture: web of data Multi-scale relations Network relations HTTP HTTP HTTP Web of data Land Use data set Climate data set HTTP HTTP The web of data layer links the cellular space to existing datasets, like Geonames, DBPedia and SWEET Ontology.
  28. 28. Web of data layer Land Use Dataset Climate Dataset Social Behaviour Dataset DBCells GeoNames DBPedia .... .... .... SWEET Ontology
  29. 29. Dbcells dataset Publication layer Architecture: Data access and storage layer Multi-scale relations Network relations HTTP HTTP HTTP Web of data Land Use data set Climate data set HTTP HTTP Integration web and local data Data access and storage layer HTTP The data access and storage layer integrate local and web data, providing a transparent access and storage for modeling tools.
  30. 30. Dbcells dataset Publication layer Architecture: model and user layers Multi-scale relations Network relations HTTP HTTP HTTP Web of data Land Use data set Climate data set HTTP HTTP Integration web and local data Data access and storage layer HTTP Model A Model layer Model B Model C Model D User layer The model layer uses and shares data provided by the lower layers. The user layer runs and reproduces the experiment of a particular model.
  31. 31. Dbcells dataset Publication layer Architecture Multi-scale relations Network relations HTTP HTTP HTTP Web of data Land Use data set Climate data set HTTP HTTP Integration web and local data Data access and storage layer HTTP Model A Model layer Model B Model C Model D User layer
  32. 32. Challenges ● ● ● ● ●
  33. 33. Final remarks
  34. 34. Final remarks
  35. 35. References
  36. 36. Thank you

×