Release Notes - PoolParty 5.1

Here you can find the release notes for PoolParty release 5.1. In the release notes you can find major developments, improvements and changes made in the respective release. The release notes are divided in three chapters:



Updates of 3rd party libraries

Release 5.1 includes several updates of 3rd party components used by the PoolParty server. The following list provides an overview:

  • Update of Apache Tomcat to version 7.0.61
  • Update of Sesame to version 2.7.15
    This update provides:
    •  full SPARQL 1.1 compatibility.
    • additional RDF serialization formats (N-Quads, RDF/JSON)
  • Update of Apache Solr to version 4.10.4
  • Update of Apache Tika to version 1.7

Security improvements

Release 5.1. includes several security improvements for the PoolParty server:

  • A default error page is displayed for all PoolParty web applications if a non-existing directory/page is requested.
  • PoolParty has been secured against Cross-Site Request Forgery (CSRF) attacks.
  • PoolParty has been secured against Cross Site Scripting attacks.


GraphSearch for Projects 

PoolParty offers the ability to calculate a GraphSearch per project based on the calculated corpora in corpus management. Every corpus can be added to GraphSearch. The calculation of the GraphSearch index is in this case done as a separate step at the end of a corpus analysis.

After successful calculation, a GraphSearch interface is available via a persistent URL (http://<server-URL>/PoolParty/<projectId>/graphsearch). Like with the wiki frontend, for public projects the Graph Search is publicly available. In the GraphSearch interface you can search over the documents. You can filter by source (corpus) and also by Facets based on the structure of the thesaurus.


PoolParty Snapshot Dashboard

The Admin Dashboard has been replaced by the PoolParty Snapshot Dashboard that is available directly as an additional view in the UI. In the Dashboard you can switch between snapshots for projects and system repositories (e.g. custom schemes). All generated snapshots are available can be downloaded or restored. In addition manual snapshots can be triggered and a note can be added to them. For project snapshots the autosave interval can be changed.

Snapshot Dashboard

Publishing of custom schemes

Similar to the linked data frontend for projects a scheme publishing functionality for custom schemes has been added. A human readable version of the scheme is displayed per default when accessing the scheme URL in a browser.

Published custom scheme


Custom scheme classes returned in Extractor

PoolParty Extractor now returns custom class information for concepts in the extract call.



Unified URI management

The management of URI creation in PoolParty has been aligned for projects and custom schemes. For both a URI pattern can be defined based on the baseURL an identifier where the identifier may include slashes. While ID generation for the individual instances for projects has not changed for custom schemes it can be defined if hash (#) or slash ( / ) URIs are generated. The Identifier for classes, resources and attributes can be defined individually and may include slashes. That way the following patterns are possible for e.g.

  • projects: http://<projectbaseURL>/ns/project/<conceptID>
  • class: http://<schemebaseURL>/Marketing/Person/Athlet


URI pattern definition for custom classes

Additional Improvements

  • Merging concepts custom scheme definitions are taken into account.
  • The Linked Data Harvesting and the Autopopulate feature have been aligned to provide the same functionality
  • The RDF Project Export dialogue has been restructured to provide better usability.
  • The session time out for PoolParty can now be set via a setting in the file.
  • The lock functionality in the Extracted Terms list, the Candidate Terms and the Suggested Concepts list has been made more visible and a text has been added.
  • In the PoolParty details view relations and attributes are displayed in different colors so they can be distinguished visually.
  • Document upload in Corpus Management skips documents that throw an error during upload.


Improved thesaurus based disambiguation

Thesaurus based disambiguation has been improved. Now more contextual information is taken into account and especially for larger thesauri with higher number of ambiguous concepts extraction precision and speed of creation of the extraction model has been improved.


  • The Visual Browser is now also available on project level.
  • Projects with multiple languages can be imported also if another projects exists with the same title in a non-default language.
  • Custom classes are inherited correctly creating concepts via the autopopulate feature.
  • Custom classes, relations or attributes added/linked to a custom scheme from a existing ontology are handled correctly during Excel import.
  • The user that did the last edit for a concept is added as a dcterms:contributor.