Deletes triples in a Semantic Web endpoint based on the triples from the input table. The triples in the table have to match the triples in the endpoint exactly.

In order to prepend namespaces it is necessary to choose Prepend in the settings dialogue.

If all triple columns are selected (i.e. subject, predicate, object column), the table is read out row-wise, so exactly the given triples are deleted. It is possible to leave out subject, predicate, object cells empty; but at least one cell as to be non-empty. For instance, the following entries for triples are possible:

Subject Predicate Object
Peter Country
Peter Germany
Peter Country Germany
Country Germany

If one or two of the triple columns are selected, only the selected columns will be read-out. For instance, if the input table is like
Subject Predicate Object
Peter Country Germany
but only subject and object column is selected, the triple will be interpreted as
Subject Predicate Object
Peter Germany

If only "graph name" column is selected, corresponding graph including stored triples, will be deleted. Attention: If all triples in a graph are deleted, the graph can stil exists (depends on connected endpoint). To be sure that a graph is completely deleted, it has to be deleted by selecting only the "Graph name" column.

Please see node description of Semantic Web node "SPARQL Insert" for usage of options "Graph name" and "Global Graph name".

Functionality is provided using the Apache Jena Framework

Input Ports

  1. Type: Semantic Web Connection Semantic Web connection
  2. Type: Data Table containing triples and graph names

Output Ports

  1. Type: Semantic Web Connection Semantic Web connection

Find here

KNIME Labs > Semantic Web/Linked Data > Read/Write

Make sure to have this extension installed:

KNIME Semantic Web

Update site for KNIME Analytics Platform 3.7:
KNIME Analytics Platform 3.7 Update Site

How to install extensions