Web Interface to the wimuQ Service

Enter your SPARQL query in the textbox below.

Sample queries:
  1. Federated query from FEDBench (with non-dereferenceable URIs)
  2. HDT 2(non-dereferenceable URI)
  3. HDT 3(URI dereferenceable)
  4. WIMU CBD (non-dereferenceable URI, Creating CBD from WIMU dumps)
  5. HDT 3
  6. SQUIN 1 (URI derefenceable)
  7. SQUIN 2 (URI dereferenceable)
  8. HDT 4

Preferred Result Format: JSON
Ignore query result cache
Execute all sample queries
Execute only on WimuTraversal
Execute only on SQUIN
Include Datasets from ReLOD

Upload your own dataset !.

Learn more about SQUIN on the SQUIN website.

Learn more about WIMU on the WIMU website.

What this SPARQL query engine can do:

The following figure shows the workflow of the query processing in WimuQ, which comprises of four main steps: (1) the user issue a SPARQL query to the WimuQ interface, which (2) extracts all the URIs used in the given user query. Note the URIs can be used in subject, predicate, or objects of the SPARQL triple patterns. (3) The extracted URIs are then searched in the WIMU index, which gives all the relevant datasets where the extracted URIs can be found. (4) The relevant datasets are furthered filtered based on the source selection algorithm and (5) the finally-selected relevant datasets are then queried by using the different query processors, depending on the type (HDT, endpoint, datadump, dereferenceable dataset) of the datasets; (6) the results generated by the different query processors are then integrated and sent back to the user. The whole process is like a black box to the end user: the user only sends the query and get back the results without knowing the underlying query execution steps.

More details in the paper, accepted at K-CAP 2019:

More Complete Resultset Retrieval from Large Heterogeneous RDF Sources. Tenth International Conference on Knowledge Capture(K-CAP). November, 2019 Marina del Rey, California.