Version 5.0.9 includes UMBEL Class Lookups and Named Entity Extraction
I first wrote about OpenLink Software‘s stellar suite of structured Web-related software back in April 2007, with a spotlight on Virtuoso, the company’s flagship ‘universal server’ product. As it has for years, OpenLink continues a steady drumbeat of new releases and extensions. The most recent version upgrade, 5.0.9, was announced today.
In the intervening period I have now personally had the chance to experience Virtuoso first hand, both as the standard hosting platform for Zitgist’s linked data products and services, and as the hosting environment for UMBEL‘s various and growing Web services. I can state quite categorically that our ability to get things done fast with few resources depends critically on the unbelievable high-productivity platform that Virtuoso provides. (And, hehe, given our close relationship to OpenLink, we also get great responsiveness and technical support! Though, truthfully, OpenLink continues to amaze with its outreach and embrace of all of the important initiatives within the semantic Web community.)
I normally let these standard Virtuoso release announcements pass without comment. But today’s release v. 5.0.9 has an especially important feature from my parochial perspective: the first support for UMBEL.
Just to refresh memories, OpenLink’s Virtuoso is a cross-platform universal server for SQL, XML, and RDF data, including data management. It includes a powerful virtual database engine, full-text indexing, native hosting of existing applications, Web Services (WS*) deployment platform, Web application server, and bridges to numerous existing programming languages. Now in version 5.0, Virtuoso is also offered in an open source version. The basic technical architecture of Virtuoso and its robust capabilities is:
From an RDF and linked data perspective, Virtuoso is the most scalable and fastest platform on the market. Critically from Zitgist’s perspective is Virtuoso’s more than 100 built-in RDF-izers (or “Sponger cartridges”) that address all major data formats, serializations, relational data and Web 2.0 APIs. But don’t take my word for it: Check out OpenLink’s impressive list of these cartridges and their various linkages throughout the linked data space.
The key aspect of the new UMBEL support in Virtuoso is its incorporation of UMBEL lookups and its use of Named Entity extraction into the RDF-izer cartridges. This is but the first of growing support anticipated for UMBEL.
Other New Features
In addition to UMBEL, this version 5.0.9 includes significant performance optimizations to the SQL Engine, SPARQL+RDF Engine, and the ODBC and JDBC drivers.
Other new features include:
- An Excel mime-type output option in the SPARQL endpoint
- Enhanced triple options for bif:contains plus new options for transitivity
- New RDF-izer Cartridges for the Sponger RDF Middleware Layer
- Support for very large HTTP client requests
- A sparql-auth endpoint with digest authentication for using SPARUL via SPARQL Protocol
- New commands for the Ubiquity Firefox plugin.
Finally, per usual, there are also minor bug-fixes:
- Memory leaks
- SQL query syntax handling
- SPARQL ‘select distinct’
For More Details
For more details, you can see these Virtuoso release notes: https://sourceforge.net/project/shownotes.php?release_id=626647&group_id=161622