Catalogue entries - KarinBecker/LODCubeSurvey GitHub Wiki

The two files below contain the Mannheim Catalogue entries that were extracted using a crawler. The tags used were "format-cube" and "governamental". As of late March 2015, there were 114 entries. The two files bellow indicate the catalogue entry, the sparql endpoint (or alternative method), graph names, and whether the entry was fully operational (possibly with the workarounds).

[Open community entries] (https://github.com/KarinBecker/LODCubeSurvey/blob/master/LODExtraction/OpenCommunityEntries.xlsx). These correspond to 83 entries of the catalogue, all of them corresponding to the sparql enpoint http://opendatacommunities.org/sparql. We extracted 65 unique graph names from these entries. In our extractor, these graphnames need to be placed in the file 'graph.txt' (provided together with our source code).

[Non Open community entries] (https://github.com/KarinBecker/LODCubeSurvey/blob/master/LODExtraction/NonOpenCommunityEntries.xlsx). These correspond to all other entries of the catalogue. Some endpoints were no longer operational (6), and some we did not managed to find workarounds for issues.
In our extractor, these endpoints need to be placed in the file 'endpointsandlinks.txt' (provided together with our source code) according to the following convention:

• odd line: sparql endpoint (e.g 1 worldbank.270a.info/sparql)
• even line: graph name (e.g. 2 http://worldbank.270a.info/graph/void). We did not face this situation, but in the future, if there is an enpoint without graph, a blank line should be inserted.