kcoyle
commented
1 year ago This would require something other than dcterms:subject and loc/authorities, since those would also be used for specific subject headings. This could be a matter of practice, using something like one of the classification systems (e.g. DDC or UDC) for the field of work, and the subject headings for the more specific terms. The classification systems have a more identifiable hierarchy than the subject headings so it would be possible to select a broad topic area. That would also be the case with DBpedia or Wikipedia topics, although those may not connect well to broader or narrower terms, making it at times difficult to find the right topic level.
The other option is to create a special property for the general field of work. That could take either a string or a URI, and I assume that the values would be different in different countries and different research areas. Would you need to translate from one academic system's terms to another, so that there was a single set of values, or could untranslated lists be used? I guess that depends on the needs for sharing.
osma
A problem with the current scholarly resource special in Latin American production is identifying the correct knowledge area of an article. On this hand, the algorithm that could be using work to associate an author or resource with a field of work is DBpedia, vocabularies such as the Library of Congress Subject Headings (LCSH) or Medical Subject Headings (MeSH), Mathematics Subject Classification (MSC) . For that, I think that we can recommend a standard for this purpose, and helping to the algorithms to have more precision to group work in a variety of fields of work. Example 1: http://example.com/resource/123 rdf:type bibo:Article ; dc:title "A Study of Machine Learning Algorithms" ; dc:creator "John Smith" ; dcterms:subject http://id.loc.gov/authorities/sh85029528 .
In this example, have a scholarly resource with a title and author that is associated with the Library of Congress Subject Heading (LCSH) identifier "sh85029528", which represents the discipline of Computer Science. This is indicated using the dcterms:subject property, which is used to specify a subject or topic for the resource. The rdf:type property is used to indicate that this resource is an article, and the dc:title and dc:creator properties are used to provide the title and author of the resource, respectively.
Example 2: http://example.com/resource/123 rdf:type bibo:Article ; dc:title "A Study of Machine Learning Algorithms" ; dc:creator "John Smith" ; dbpedia-owl:wikiPageWikiLink http://dbpedia.org/resource/Computer_science .
To associate this resource with the field of Computer Science, use the DBpedia resource identifier for Computer Science (http://dbpedia.org/resource/Computer_science) and link it to the resource using the dbpedia-owl:wikiPageWikiLink property, as shown in the previous example.