Characterizing the Semantic Web on the Web

Presented at: 5th International Semantic Web Conference (ISWC2006)

by Li Ding, Tim Finin


The applicability of the Semantic Web in real world applications on the World Wide Web remains unclear because most existing work neglects the Web aspect: Semantic Web data is published and consumed on the Web by independent agents. Therefore, we developed a system to harvest and analyze publicly available Semantic Web data on the Web. We estimate at least 10 millions of Semantic Web documents on the Web using Google, and we have actually harvested 1.4 million documents comprising over 250 million triples. Based on the collection, we measured a rich spectrum of global properties and usage patterns of Semantic Web documents and terms. The most interesting observation is that the distribution of many metrics, such as the size of Semantic Web documents and the meta-usages of Semantic Web terms, follow the Power law.

Resource URI on the dog food server:

Explore this resource elsewhere: