Upload File Instructions

Each user can upload a csv file for visualizing any set of sources. The first line of the csv file should be the following:
Source,Category,Triples,URIs,Literals,Blank_Nodes,Degree

Then, each line represents one source. For each source the user should at least give the name, the category and the number of triples
and can optionally fill the other features.
The categories are the following:

The number of triples and the other features can be found from the following SPARQL queries:
Feature SPARQL Query
# of Triples SELECT count(*) as ?numberOfTriples from <Source> where { ?s ?p ?o}
# of URIs SELECT (count (distinct ?s) as ?numberOfURIs) from <Source> where{ { ?s ?p ?o } UNION { ?o ?p ?s } FILTER(isURI(?s)) }
# of Literals SELECT (count (distinct ?o) as ?numberOfLiterals) from <Source> { ?s ?p ?o FILTER(isLiteral(?o)) }
# of Blank Nodes SELECT (count (distinct ?s) as ?numberOfBNodes) from <Source> { { ?s ?p ?o } UNION { ?o ?p ?s } FILTER(isBlank(?s)) }
Degree of URIs SELECT xsd:double(count(?incomingProperties)+count(?outgoingProperties))/xsd:double(count (distinct ?node)) as ?degreeOfURIs from <Source>
where{ {?node ?outgoingProperties ?o } UNION {?o1 ?incomingProperties ?node} . FILTER(isURI(?node) )}

After finishing with the sources, one can write the links existing between the pairs of sources.
The two lines after the last source should always be the following:
#links
source1,source2,commonURIs

Afterwards, the user can write how many links exist between any pair of sources.
Each pair of sources (having at least one common link) is written once because we assume that the links are undirected.
Note that the last line of the file must not be an empty line!
A whole example is given in this sample csv file.


Back to manual