a multilingual classification system for cultural content

ICONCLASS as Linked Open Data

Every notation in ICONCLASS is associated with a Uniform Resource Identifier (URI) that is unique to that subject heading. For example the concept "Communication of Thought" has a notation of 52D1 and thus the URI becomes: http://iconclass.org/52D1 If you click on the aforementioned link, you are redirected to a HTML representation of that concept, fit for human consumption.

It is also possible to retrieve machine-readable formats, by either appending '.rdf' or '.json' to the URI. This will redirect you to either a SKOS/RDF or JSON representation of the concept. For example http://iconclass.org/52D1.rdf or http://iconclass.org/52D1.json

If you need to retrieve the data for notations containing 'keys' like a 25F23(+46) sleeping lion ensure that you properly encode the + sign in the URL. The link becomes: http://iconclass.org/25F23(LION)(%2B46).json Note how the + sign has been replaced with the characters %2B For more information have a look at the Wikipedia article for Percent Encoding. You have to do the same for names, like: 11H(BASIL THE GREAT)


To search for notations using natural language you simply need to add the parameter: 'fmt=json' to any search. For example: http://iconclass.org/rkd/0/?q=lion&q_s=1&fmt=json

Retrieving texts for multiple notations in one HTTP call

You can get the text that go with a list of notations in one go, by calling the URL /json/ or /rdf/ with a list of 'notation' parameters for which you would like the data. Here is an example: http://iconclass.org/json/?notation=25FF&notation=51A

Do you want the raw data?

Please don't start scraping away, or fire up the scripts. There are about 1 million notations in ICONCLASS (with all keys and children full expanded).

There is an updated complete dump of the ICONCLASS database in SKOS JSON-LD NDJSON available at: http://iconclass.org/data/iconclass_20200710_skos_jsonld.ndjson.gz(MD5 of the file is 4847746c85bedea87cedc6e6f0643952 )

Note that this data file is about 140MB compressed with GZip, and when expanded is more than 2GB.

If you like working with raw data, it might be better to download the data from Github, see: https://github.com/epoz/iconclass and an application of ICONCLASS plus a Python library at: https://labs.brill.com/ictestset/

Page last edited on: 28 May 2020

Mail: eposthumus@gmail.com


ICONCLASS is made available under the Open Database License: http://opendatacommons.org/licenses/odbl/1.0/. Any rights in individual contents of the database are licensed under the Database Contents License: http://opendatacommons.org/licenses/dbcl/1.0/