Tagged: Querying Wikidata
- September 12, 2016 at 1:55 pm #68787CorfromleuvenMember
I heard your presentation about importing data in Nodegoat in Mons earlier this year (https://nodegoat.net/blog.p/82.m/16/members-of-the-us-house-of-representatives—wikidata )
I’am interested in doing the same for lists of European Nobility between 1800 & 1815.
These list exist, for example like these:
Making me wonder that they might be based on queries in wikidata already.
It would be of great help to look even further for any other attributes connected to these names (place of birth etc.)
I’am currently trying to identify a lots of these people figuring in lists with invitees for public festivities in Belgium. It would be of great help to be able to import them with as many attributes possible.
I you can show me how this could be done, I could do most of the work myself of course.
The aim is very much to share the results with the nodegoat community of course, so that they can be used as linked open data for other projects.
Thanx in advanceSeptember 12, 2016 at 3:10 pm #68791nodegoatKeymaster
As you mentioned yourself, wikidata would be the best place to go, as you can find a lot of this data there in a structured format. We just checked and with a simple SPARQL query, we were able to get 3864 results (click run to load the results).
You can check this blogpost: https://nodegoat.net/blog.s/12/linked-data-vs-curation-island, this tutorial: http://programminghistorian.org/lessons/graph-databases-and-SPARQL, and these examples: https://www.mediawiki.org/wiki/Wikibase/Indexing/SPARQL_Query_Examples to learn how you can include more attributes in your results.
Once you have the results you want, you just download them as CSV and use the steps described here to import them into nodegoat: http://historicalnetworkresearch.org/forums/topic/import-functionality/ . In case you want to import relational data, it might be good to do this in multiple steps (so first import all the titles into a classification, and then import persons with an object description ‘title’ that links to this classification).
Does this help you?September 13, 2016 at 10:42 am #68792CorfromleuvenMember
No it doesn’t. Graf datanbases and SPARQL are useless unless you:
a) have several weeks of time to spare
b) receive a decent “map” of how the data are stored in de database (in which case you will probable save some days)
The examples of British Museum query’s in the tutorial prove that: you just can’t now in advance where the date fields are stored, unless they show it in a clearly readable graph.
You can of course always find out by trial and error (it is an open system after all) and a bit of logical thinking.
But then you need several weeks to find out.
So this is of no use for me at this stage of my research.
And it is not about logic, believe me. It is about the time to find out because of the lack of clear documentation that shows you how it really works.
Wikidata is similar in that respect: no usable documentation available. It’s a shame, but not my problem. I will check it out anyhow one by one. For my case this is sufficient, but it doesn’t add up to the usefulness of your platform I guess.September 13, 2016 at 1:50 pm #68793nodegoatKeymaster
Sorry to hear that. Could you perhaps say a bit more about the sort of help you are looking for?
Your goal is to generate lists of people from the nobility with as many attributes as possible, and you want to use the data you mentioned in your original post, or did we misinterpret your question? Are you looking for something like this: http://wikitables.geeksta.net/url/?url=https%3A%2F%2Ffr.wikipedia.org%2Fwiki%2FListe_des_membres_de_la_noblesse_d%27Empire ?
You are right that documentation is lacking, but the beauty of these document/graph databases is also that they are quite self-documenting. If you click the top result of the query we made, Francis d’Allarde (https://www.wikidata.org/wiki/Q15980290), you see a number of relevant statements/predicates that you can use for all the other persons as well: date of birth (P569), date of death (P570), sex or gender (P21), country of citizenship (P27), place of burial (P119) etc. etc.
We’re happy to assist you further with this!
- You must be logged in to reply to this topic.