Skip to content

Knowledge Graph¤

The Knowledge Graph plugin is a dataset for reading and writing RDF to a knowledge graph embedded in Corporate Memory.

Description¤

The Knowledge Graph plugin has the technical ID eccencaDataPlatform. This ID contains a reference to eccenca Explore, also known as DataPlatform.

Conceptually, the main responsibility of this plugin is to be a RDF dataset for reading and writing RDF statements into or from a knowledge graph. This knowledge graph is embedded within eccenca Corporate Memory (CMEM), specifically within the already mentioned DataPlatform, known as eccenca Explore.

The plugin itself is part of eccenca Build (internally known as DataIntegration). In a nutshell, the relevant difference between Build and Explore is the following aspect (see the Getting Started and the remaining documentation for further details):

  • Explore is for browsing and exploring knowledge graphs
  • Build is for creating and integrating knowledge graphs

This plugin, i.e. eccencaDataPlatform, is thus responsible for creating and integrating knowledge graphs within eccenca Build. The knowledge graph itself is part of DataPlatform, which can be explored in the UI of eccenca CMEM under the Explore module.

Example usage¤

The usage of this plugin is showcased, for example, in these three tutorials, where the knowledge graph is used as a sink dataset. The creation of the knowledge graph is handled implicitly.

Another tutorial showcasing how to consume the knowledge graph, i.e. how to use it as a source of information rather than as a sink, is this example with a SQL database as a source dataset.

Other types of RDF datasets are the in-memory dataset and the RDF dataset. These are worth considering if the information is short-lived or the dataset is small. A more durable and resilient solution is to use an Enterprise Knowledge Graph, which is exactly what this plugin does.

Next to the knowledge graph embedded within eccenca CMEM, there is the possibility to use an external SPARQL endpoint. The embedded knowledge graph of CMEM is an essential component of CMEM as semantic data management software. The additional possibility of integrating with external knowledge graphs via the SPARQL endpoint is merely a small part of the possibilities to consume data within CMEM.

Parameter¤

Graph¤

The URI of the named graph.

  • ID: graph
  • Datatype: graph uri
  • Default Value: None

Clear graph before workflow execution¤

If set to true, this will clear the specified graph before executing a workflow that writes to it. Note that this will always use the configured graph and ignore any overwritten values from the config port.

  • ID: clearGraphBeforeExecution
  • Datatype: boolean
  • Default Value: false

SPARQL query timeout (ms)¤

SPARQL query timeout in milliseconds. By default, a value of zero is used. This zero value has a symbolic character: it means that the timeout of SPARQL select and update queries is configured via the properties silk.remoteSparqlEndpoint.defaults.connection.timeout.ms andsilk.remoteSparqlEndpoint.defaults.read.timeout.ms` for the default connection and read timeouts. To overwrite these configured values, specify a (common) timeout greater than zero milliseconds.

  • ID: sparqlTimeout
  • Datatype: int
  • Default Value: 0

Optimized entity retrieval¤

Optimized retrieval method to remove load from the underlying triple store. Query parallelism is limited and cheaper queries are executed against the backend. By putting the main work on DataIntegration side, the RDF backend is kept responsive.

  • ID: optimizedRetrieve
  • Datatype: boolean
  • Default Value: true

Advanced Parameter¤

Endpoint¤

The named endpoint within eccenca DataPlatform.

  • ID: endpoint
  • Datatype: string
  • Default Value: default

Page size¤

The number of entities to be retrieved per SPARQL query. This is the page size used when paging.

  • ID: pageSize
  • Datatype: int
  • Default Value: 100000

Pause time¤

The number of milliseconds to wait between subsequent query

  • ID: pauseTime
  • Datatype: int
  • Default Value: 0

Retry count¤

The number of retries if a query fails

  • ID: retryCount
  • Datatype: int
  • Default Value: 3

Retry pause¤

The number of milliseconds to wait until a failed query is retried.

  • ID: retryPause
  • Datatype: int
  • Default Value: 1000

Strategy¤

The strategy for retrieving entities. There are three options: simple retrieves all entities using a single query; subQuery also uses a single query, which is optimized for Virtuoso; parallel executes multiple queries in parallel, one for each entity property.

  • ID: strategy
  • Datatype: enumeration
  • Default Value: parallel

Entity list¤

A list of entities to be retrieved. If not given, all entities will be retrieved. Multiple entities are separated by whitespace.

  • ID: entityList
  • Datatype: multiline string
  • Default Value: None

Comments