Wikidata:Property proposal/OpenAlex ID

OpenAlex ID edit

Originally proposed at Wikidata:Property proposal/Generic

Done: OpenAlex ID (P10283) (Talk and documentation)

Description	Identifier for works, authors, institutes, venues, concepts/subjects in OpenAlex (Q107507571).
Data type	External identifier
Domain	item
Allowed values	[WAVIC][1-9]\d{3,9} (the real identifiers are not case-sensitive, but in Wikidata they will be upper case only, like with the DOI)
Example 1	The state of OA: a large-scale analysis of the prevalence and impact of Open Access articles (Q49510981) → W2741809807 (a work)
Example 2	Jason Priem (Q21678556) → A2208157607 (an author)
Example 3	Journal of the Welsh Bibliographical Society (Q6296212) → V87782555 (a venue)
Example 4	University of North Carolina at Chapel Hill (Q192334) → I114027177 (an institute): compare with Microsoft Academic ID (P6366) = 114027177
Example 5	altmetrics (Q14565201) → C2778407487 (a concept): compare with Microsoft Academic ID (P6366) = 2778407487
Source	https://docs.openalex.org/entity-objects#the-openalex-id
Planned use	add OpenAlex identifiers to Wikidata, allowing federated searching
Number of IDs in source	millions
Expected completeness	always incomplete (Q21873886)
Formatter URL	https://openalex.org/$1
See also	Microsoft Academic ID (P6366)

Motivation edit

OpenAlex (Q107507571) is a CC0 licensed resource from the team that brought us Unpaywall and generally a key player in the open knowledge field. The OpenAlex ID for concepts are already mapped to Wikidata pages. Egon Willighagen (talk) 15:40, 4 January 2022 (UTC)[reply]

Some note the Microsoft Academic ID (P6366) and the initial OpenAlex ID identifiers follow the Microsoft Academic ID (P6366). But MAG is only one source and it is not a 1-to-1 match, MAG is discontinued, and the OpenAlex team already curated data. --Egon Willighagen (talk) 07:29, 10 January 2022 (UTC)[reply]

Discussion edit

Support looks useful. TiagoLubiana (talk) 17:00, 4 January 2022 (UTC)[reply]
Support no brainer. --DarTar (talk) 17:02, 4 January 2022 (UTC)[reply]
Support cool! ArthurPSmith (talk) 18:20, 4 January 2022 (UTC)[reply]
Support Clearly useful. --Daniel Mietchen (talk) 16:38, 5 January 2022 (UTC)[reply]
Comment if there are clearly separate ranges of values, I'd probably consider separate properties. --- Jura 10:52, 6 January 2022 (UTC)[reply]
I intentionally did not do that. When I proposed exactly this with KEGG identifiers, it was deemed unwanted. Egon Willighagen (talk) 17:11, 7 January 2022 (UTC)[reply]
I don't recall the details of KEGG (despite being mentioned on its talk page), but currently it doesn't seem to have any complex constraints defined based on value ranges so it might not have been that important. Also, the numbers of entries are likely to be much higher. Do we have an estimate for this? --- Jura 12:51, 8 January 2022 (UTC)[reply]
Currently, the chemical dimension of KEGG is not huge: several tens of thousands of metabolites, drugs, and glycans but close to 39 million genes (see https://www.genome.jp/kegg/docs/statistics.html). So, higher, but not that much. --Egon Willighagen (talk) 07:35, 10 January 2022 (UTC)[reply]
I see. Apparently KEGG isn't much used then despite being available for year. So it's not clear if the comparison really helps. --- Jura 19:03, 10 January 2022 (UTC)[reply]
I do not know how you quantified the use of KEGG. Can you explain? --Egon Willighagen (talk) 15:55, 11 January 2022 (UTC)[reply]
Number of uses at WD compared to total number of ids. --- Jura 08:41, 22 January 2022 (UTC)[reply]
Support. --Vladimir Alexiev (talk) 22:36, 6 January 2022 (UTC)[reply]
- Replaces Microsoft Academic ID (P6366).
- Should be straightforward to populate from existing values by prepending a type letter: if Human then A, if Organization then I, etc. Any volunteers to check out this idea?
  - I'll check it out. I think there will be duplication between C (concept) and the other identifiers. --Nikola Tulechki (talk) 08:09, 7 January 2022 (UTC)[reply]
- OpenAlex includes some WD ids (at least for Concepts) so a useful crosscheck is possible.
- However @Egon_Willighagen: I think the prefix letter should be Uppercase. Just because OpenAlex allows either case, does not mean it's a good idea to allow it here. I propose regex [WAVIC][1-9]\d{3,9} .
  - Agree the case-sensitivity of the IDs. Wikidata is not case insensitive. It follows that if we allow that here it will generate problems (e.g distinct-values constraint (Q21502410)) --Nikola Tulechki (talk) 08:09, 7 January 2022 (UTC)[reply]
  - Replaced regex (was [WwAaVvIiCc]\d+) --Vladimir Alexiev (talk) 15:49, 7 January 2022 (UTC)[reply]
    That would be a Wikidata choice then. The identifier itself is not case sensitive, but I understood that it makes comparison in tools easier if they are always upper case, like with DOIs. --Egon Willighagen (talk) 17:08, 7 January 2022 (UTC)[reply]
Support. This prop is indispensable! Let's wish a long and colorful future for OpenAlex (Q107507571) --Nikola Tulechki (talk) 08:09, 7 January 2022 (UTC)[reply]
Oppose in the proposed format: it might easier to create separate properties for each type. Also, if one drops the initial letter from the identifier, cross-checking with Microsoft Academic ID (P6366) would be simple. --- Jura 19:03, 10 January 2022 (UTC)[reply]
Microsoft Academic is abandoned (see https://www.microsoft.com/en-us/research/project/academic/articles/microsoft-academic-to-expand-horizons-with-community-driven-approach/ which actually points people to OpenAlex as solution to use instead). There is no reason to compare it with that. --Egon Willighagen (talk) 15:00, 11 January 2022 (UTC)[reply]
We actually have plenty such statements and the data is still available.
Not doing so would be waste of volunteer contributions to Wikidata. --- Jura 12:23, 18 January 2022 (UTC)[reply]
@Jura1, sorry, not sure how to get to that conclusion, but it's not true: this proposal does not overwrite past work at all, nor suggests deleting an old identifier. --Egon Willighagen (talk) 07:20, 20 January 2022 (UTC)[reply]
Strange, above you wrote "There is no reason to compare it with that". Can you clarify your uses of this property. --- Jura 08:41, 22 January 2022 (UTC)[reply]
The use of this property is explained above: to link to OpenAlex. --Egon Willighagen (talk) 15:15, 23 January 2022 (UTC)[reply]

To summarize about the comparing, you can only compare OpenAlex ID with Wikidata, because Microsoft Academic no longer exists. This identifiers does not replace, remove, or whatever the MA identifier. OpenAlex IDs will show differences, because of the curation work they did. Are you suggesting that Wikidata trumps an external database? Where you find changes, you will have to still manually discuss with the OpenAlex team what the ground truth is; passed MA identifier work cannot resolve that situation. Therefore, no Wikidata volunteer effort is wasted, as you instigate. --Egon Willighagen (talk) 15:25, 23 January 2022 (UTC)[reply]

Support -- Oa01 (talk) 21:04, 10 January 2022 (UTC)[reply]

Support Will be very useful! AdrianoRutz (talk) 16:08, 12 January 2022 (UTC)[reply]
Support — MasterRus21thCentury (talk) 19:59, 12 January 2022 (UTC)[reply]

Tried to count type combinations of things with MAG but got timeout. Need to run this on a local instance?

select ?A ?I ?V ?W (count(*) as ?c) {
  ?x wdt:P6366 ?mag
  bind(exists {?x wdt:P31/wdt:P279* wd:Q5} as ?A)
  bind(exists {?x wdt:P31/wdt:P279* wd:Q43229} as ?I)
  bind(exists {?x wdt:P31/wdt:P279* wd:Q5633421}|| exists {?x wdt:P31/wdt:P279* wd:Q625994} as ?V)
  bind(exists {?x wdt:P31/wdt:P279* wd:Q17537576} as ?W)
} group by ?A ?I ?V ?W

Try it!

--Vladimir Alexiev (talk) 16:00, 7 January 2022 (UTC)[reply]

@Egon Willighagen, TiagoLubiana, DarTar, ArthurPSmith, Daniel Mietchen, Jura1: @Nikola Tulechki, Oa01, Vladimir Alexiev, AdrianoRutz, MasterRus21thCentury:

Done OpenAlex ID (P10283) Pamputt (talk) 16:23, 23 January 2022 (UTC)[reply]

the identifiers for the OpenAlex venues changed the format. Their first letter used to be V, now - S. The urls changed as well - it is https://explore.openalex.org/sources/S73261239 instead of https://explore.openalex.org/venues/V73261239. ( --Waydze (talk) 06:31, 10 April 2023 (UTC)[reply]