PoolParty System Architecture

PoolParty Semantic Suite is a Java based application that can run in an Apache Tomcat 7 or IBM WebSphere 8.5.5 web application server. PoolParty's Frontend is based on JavaScript and WebFrameworks.

PoolParty comes with a built-in RDF4J graph database used for managing taxonomy & thesaurus data and ontology data. PoolParty uses a built-in Apache Solr server to build entity extraction models and GraphSearch indices. Apache Spark is used to perform machine learning tasks for semantic classification of documents.

In addition to or while replacing the built-in RDF4J graph database (GDB), external graph databases can be connected via PoolParty's Semantic Middleware Configurator. PoolParty offers various features which work with external graph databases. The following matrix outlines which stores are available and which features can be used in that context.

PP Functionality Project/Ontology Export CorpusManagement Store Extraction Results (ppx annotate/store) GraphSearch GraphEditor Replace built-in GDB
Allegrograph (tick) (error) (tick)


(error)  (error) 


(tick) (error) (tick) (tick) (error)  (error) 
MarkLogic (tick) (error) (tick) (tick) (error)  (error) 
rdf4J (tick) (tick) (tick) (tick) (tick) (tick)
Stardog (tick) (tick) (tick) (tick) (tick) (tick)
Virtuoso (tick) (tick) (tick) (tick) (error)  (error) 

neo4j (tick)  (error)   (error)   (error)  (error)  (error)

 Optionally, the built-in Apache Solr server can be replaced by an external Apache Solr or Elasticsearch server. The built-in Apache Spark ML server can be substituted by an external Spark ML Cluster.