IW Meeting 2012-02-09

From Inference Web

Jump to: navigation, search

Contents

Meeting Information

Attendees

  • Deborah
  • Cynthia
  • Tim
  • Jim
  • Paulo
  • James

Agenda

  • provenance in csv2rdf4lod IPAW paper
  • 1 - fuse status update from
  • 1.b - james on use case and presentation needs
  • 2 - prov o update / needs from pml
  • 3 - paulo / pnnl status on probeit
  • 4 round the room on other news - ieee paper in - congrats!
  • other?
  • Paulo joining the connections task force

Discussions

csv2rdf4lod provenance

1) why to trust a third part aggregator 2) provide pointer back to how a particular element appears on a third party site. (not to get blamed).

Tim still needs to give a link to an example provenance in csv2rdf4lod (but he needs to revisit what's there)

Cynthia publishing in FUSE

on Mac: command-K to connect to server

turn on vpn webdav: using web browser: https://tw1.tw.rpi.edu/fuse/ on mac, finder->go->connect to server-> https://tw1.tw.rpi.edu/fuse/ username: lebot, dlm, james.mccusker, just your email id password: add "Fuse" to your username. (to your email username on the tetherless web pages - not necessarily user name on other rpi systems)

You can access the files via a remote drive:

bash-3.2$ head /Volumes/fuse/index.html  <html> <header> <title>Fuse Documents</title> <body>

FUSE Documents

windows client: http://www.bitkinex.com/webdavclient Notes on the windows client: http://en.wikipedia.org/wiki/WebDAV#Microsoft_Windows

on rpi drupal: http://tw.rpi.edu/web/project/fuse/docShare/home


provo

http://www.w3.org/2011/prov/wiki/ProvRDF

todo - group look at the page above


Probeit   nick and hugo working intensely  there is an interface between webprobe and json

paulo

CAT system at PNNL

Colalb analysis toolkit

Paulo interested in making the data RDF.

csv2rdf4lod takes tabular data and creates RDF.

TODO: tim to send paragraph to paulo why transforming to RDF is good (by integrating it) and including provenance to give transparency.

TODO: Paulo to give tim URL of datafile with his "gridded data" example.


use case

This is the (spreadsheet-type) dataset:

http://rio.cs.utep.edu/ciserver/ciprojects/GravityMapPML/05206658643238425_gravityDataset.txt


and this is the map that was generated from that dataset:

http://rio.cs.utep.edu/ciserver/ciprojects/GravityMapPML/0799380807550336_gravityContourMap.ps

The provenance question that I have in mind is the following:

If a scientist selects a point in the map, how can I go back to the map and identifies the entries that were used to plot that select piece of the map?

converting to rdf

The following query against http://logd.tw.rpi.edu/sparql will get you started on the RDF version of your data.

prefix wgs: <http://www.w3.org/2003/01/geo/wgs84_pos#>
PREFIX conversion: <http://purl.org/twc/vocab/conversion/>
SELECT ?lat ?lon
WHERE {
  GRAPH <http://logd.tw.rpi.edu/source/trust-utep-edu/dataset/05206658643238425_gravityDataset/version/2010-Jun-08>  {
    [] wgs:lat ?lat ; wgs:lon ?lon .
  }
} 


There's provenance in the GRAPH above. That plus

http://logd.tw.rpi.edu/source/trust-utep-edu/file/05206658643238425_gravityDataset/version/2010-Jun-08/conversion/trust-utep-edu-05206658643238425_gravityDataset-2010-Jun-08.void.ttl

should be all the provenance I have.



You can recreate the RDF production by running https://scm.escience.rpi.edu/svn/public/logd-csv2rdf4lod/data/source/trust-utep-edu/05206658643238425_gravityDataset/version/retrieve.sh

Facts about IW Meeting 2012-02-09RDF feed
Date9 February 2012  +
Personal tools
Navigation