Jena-based Components for Building Semantic Web Applications

From lotico
Revision as of 12:30, 15 June 2020 by Marco (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Speaker: Claus Stadler

Location: Recording now available on Youtube (Was live online on zoom)

Date: 2 April 2020 - 6pm CEST / 5pm BST / 12pm EDT

Participants: 23


The basics of RDF and SPARQL may seem simple enough at first, but once one starts to develop a prototype, one quickly stumbles upon the same recurrent problems, such as:

How can performance be improved by means of query caching? How can we deal with blank nodes? How to deal with result set limits?

Apache Jena ( is a powerful Semantic Web toolkit, however for the aforementioned issues it does not provide out-of-the box solutions. Our original motivation for our independent "jena-sparql-api" project was to address these issues in one central place, instead of distributing solutions - possibly with various degrees of quality - among our applications (e.g. RDFUnit, DL-Learner, LIMES). By now, the library has grown. Most notably, it now features declarative Java-RDF mappings and RDF processing with reactive streams.


Claus Stadler is a researcher at the Institute for Applied Informatics / University Leipzig.

He developed an early version of the DBpedia live extraction, and was lead developer for LinkedGeoData, Sparqlify and "Semmap". The latter is a SPARQL-based faceted search application deployed at the European Open Data portal. He is also a contributor to the Big Data project "Semantic Analytics Stack" (SANSA). In his 10 years of Semantic Web experience, he worked on several funded projects, namely LOD2, LATC, GeoKnow, QROWD, HOBBIT and LIMBO. His ambitions include the exploration of approaches that simplify RDF-based data integration; process- and performance-wise. This includes design and implementations of:

(a) concepts for virtualization of RDF models, such as by means of SPARQL statement rewriting

(b) processes based on splitting monolithic RDF graphs into 'record-sized' ones and

(c) faceted search tooling.

Session-Type: Technology - Application - Coding
Session-Level: Intermediate - Advanced