Issue 553: Knowledge extraction from texts

Starting Date: 

Posted by Martin on 21/9/2021

Dear All,

Dominic Oldman and I would like to raise a new issue:

Whereas large efforts concentrate on migrating legacy data into CRM format, and the CRM is not meant to prescribe what to document, guide-lines and incentives for an optimal use of the ability of the CRM to create and cross-correlate contextual data might have a positive impact on the practice CH documentation, including mehods such as knowledge extraction from texts etc.

Posted by Dominic Oldman on 22/9/2021

Dear All,

Just adding to Martin's note.

The current CIDOC CRM says

“The CIDOC CRM is specifically intended to cover contextual information: the historical, geographical and theoretical background that gives museum collections much of their cultural significance and value”.

The practice in which organisations simply transfer their existing data to a new technology/standard without reviewing its contextual content creates several issues in terms of the contextual integration of data, communicating significance and relevance to wider audiences through data, and connected to this, and as we have discussed at the first meeting of the CRM Equality and Respect meeting, perpetuating inequality of representation in data. This is particularly important because many CH organisations directly publish their documentation records to the public as part of audience communication.

We propose a specific and regular CRM SIG discussion about how to promote the contextualising principles of the CRM - since this is a core aspect of the CIDOC CRM - and address these issues by improving and disseminating information about the relevance of the CIDOC-CRM for organisations beyond existing documentation scopes.