NanJing Vocabulary Repository (NJVR)


1 Introduction

NJVR is a vocabulary repository extracted from the index of the Falcons search engine, maintained by the Websoft Research Group at Nanjing University. It is devoted to providing the research community a representative sample of the real Semantic Web for conducting experiments and evaluations about all kinds of vocabulary-oriented research problems.

This version of NJVR contains 1,763 vocabularies (a.k.a. ontologies), comprising a total of 61,349 terms (i.e. classes and properties).


2 Downloads

NJVR consists of two TSV files and a ZIP archive of RDF documents.

2.1 Vocabularies

Each line of vocabulary.tsv provides the following tab-separated information about a vocabulary:

  1. ID
  2. Namespace URI
  3. Number of classes
  4. Number of properties

2.2 Terms

Each line of term.tsv provides the following tab-separated information about a term:

  1. ID
  2. URI
  3. Is a class (1) or not (0)
  4. Is a property (1) or not (0)
  5. ID of the vocabulary it belongs to, to be used with vocabulary.tsv (see above)
  6. URL of the RDF document obtained by dereferencing its URI
  7. ID of the RDF document obtained by dereferencing its URI, to be used with documents.zip (see below. Please note that, very occasionally due to an unsolved issue, the document is not included in documents.zip or is not a well-formed RDF document.)

2.3 RDF documents obtained by dereferencing term URIs

RDF documents in documents.zip are named DOCUMENT_ID.rdf.


3 How to cite

Gong Cheng, Min Liu, Yuzhong Qu. NJVR: The NanJing Vocabulary Repository. In Proceedings of the Joint Conference of the 6th Chinese Semantic Web Symposium and the 1st Chinese Web Science Conference, pages 265--272, 2013.


4 Earlier versions