Yes RDF is All Well and Good But Does It Scale?

A criticism of Linked Data, RDF and the Semantic Web in general is that is doesn’t scale. In the past this has been a justified complaint - but no longer.

The EU-funded Big Data Europe project, in which W3C is pleased to be a partner, is running a series of pilots in 7 different societal areas (the European Commission’s Societal Challenges). Not all of these use Linked Data but those that do are using a lot of it. OpenPHACTS, for example, provides an API to a service that reconciles the many different identifiers used in biomedical and pharmacological data and easily handles billions of triples. In the food and agriculture domain, we’re using NLP to extract information from millions of scholarly articles about viticulture and linking that information using a SKOS thesaurus. In the social sciences, Linked Data is again being used to compare spending data from neighboring authorities. Finally, we’re using large scale geospatial Linked Data to process Earth Observation data to detect change on the ground and link it to events as reported in social media.

All of these pilots depend on processing large amounts of data in a variety of formats. As I noted recently, the Big Data Europe platform creates a virtual RDF graph of all the data available at query time, overcoming the most difficult of problems with big data: variety. Allied to this use of Semantic Technologies, the team behind the Semantic Analysis Stack, SANSA, has just released version 0.2 of its distributed computing software that supports machine learning, inference and querying at scale.

This kind of advanced use of Semantic Technologies will be in focus later this year at the International Semantic Web Conference (ISWC) in Vienna, 21 – 25 October. The Big Data Europe project is proud to be a Gold sponsor of ISWC 2017, which is the premier annual forum for the discussion of Semantic Technologies and expects to attract 600-700 international delegates. As local chair Axel Polleres notes, it presents a variety of opportunities for local and international businesses. Best of all from my point of view the 16th year of this event and others like it refute the nay-sayers’ claims that the Semantic Web is an academic solution looking for a real world problem. Big Data Europe provides evidence for it being used to solve real world data processing problems at scale.