Regarding Bulk Concept import

27 views
Skip to first unread message

Arushi T

unread,
Mar 20, 2026, 9:20:50 AM (21 hours ago) Mar 20
to vocbench-user
Hello,

Based on the documentation and my experience working with the tool, I have not been able to find an API that allows me to import a set of concepts which were developed as part of a thesaurus in a mind mapping tool. Our use case is specifically oriented towards checking whether we can also use the Semantic Turkey UUID URI generator for imported concepts, and not only for concepts newly created through the UI.

Looking forward to your reply. Thank you for accepting my join request.

Best Regards,
Arushi

Thomas Francart

unread,
Mar 20, 2026, 9:29:58 AM (21 hours ago) Mar 20
to Arushi T, vocbench-user
Hello

Have you tried Sheet2Rdf in VocBench ? https://art.uniroma2.it/sheet2rdf/documentation/vb_tool/
If you can export your MindMap in a tabular spreadhseet format then you could map it in here.

I was wondering though to which extent would the Sheet2Rdf be available as an API / library to be integrated in other applications ? I am thinking of an API that would wrap Sheet2Rdf, take the Excel file + the PEARL mapping file and return the generated triples. Is this possible?

Thanks
Thomas


--
You received this message because you are subscribed to the Google Groups "vocbench-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vocbench-use...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/vocbench-user/2c0f88ba-8420-4fa0-bc93-f511142e7da4n%40googlegroups.com.


--

Thomas Francart - SPARNA
linked data | domain ontologies | knowledge graphs
blog :
blog.sparna.fr, site : sparna.fr, linkedin : fr.linkedin.com/in/thomasfrancart
tel : 
 +33 (0)6.71.11.25.97

Arushi T

unread,
Mar 20, 2026, 9:33:10 AM (21 hours ago) Mar 20
to vocbench-user
Thanks for sharing the document. I will go through the steps and get back to you. :) 

stel...@uniroma2.it

unread,
Mar 20, 2026, 9:41:49 AM (21 hours ago) Mar 20
to Arushi T, vocbench-user

Dear Arushi,

 

this depends on how you think to implement the process. As a starter, what is the output of the mind mapping tool?

 

In general, you could follow one of these approaches:

 

  1. You write a batch that reads the source info, then Invoke the APIs for concept creation, handing all the information from the mind-mapping tool. So you will use the URI generator
  2. You export the thesaurus in an spreadsheet (CSV, Excel...), then you import it with Sheet2RDF (https://art.uniroma2.it/sheet2rdf/documentation/vb_tool/)  and choose to use the URI generator
  3. You develop an implementation of the lifter extension point ( https://semanticturkey.uniroma2.it/doc/sys/rdf_lifter.jsf ) so that it is tailored to the export format of your mindmap tool. Even there, you can invoke the URI Generator through the internal API. This is the only one that requires true development, but it’s not complicated and if you are using it often, that could be a possibility.

 

Kind Regards,

 

Armando

 

--

Thomas Francart

unread,
Mar 20, 2026, 9:47:16 AM (21 hours ago) Mar 20
to vocbench-user
Hello
 
I was wondering though to which extent would the Sheet2Rdf be available as an API / library to be integrated in other applications ? I am thinking of an API that would wrap Sheet2Rdf, take the Excel file + the PEARL mapping file and return the generated triples. Is this possible?


I was trying to answer my own question, and saw that Sheet2rdf is available as a command-line tool (https://art.uniroma2.it/sheet2rdf/documentation/usage.jsf), which indicates it is autonomous from the UI. Good.
However, I was trying to have a look at the code of this CLI tool, but could not find it. Only the sheet2rdf core code is available (https://bitbucket.org/art-uniroma2/sheet2rdf/src/master/). Is this CLI tool code visible somewhere ?

Thanks !

 
Thanks
Thomas


Le ven. 20 mars 2026 à 14:20, Arushi T <arush...@gmail.com> a écrit :
Hello,

Based on the documentation and my experience working with the tool, I have not been able to find an API that allows me to import a set of concepts which were developed as part of a thesaurus in a mind mapping tool. Our use case is specifically oriented towards checking whether we can also use the Semantic Turkey UUID URI generator for imported concepts, and not only for concepts newly created through the UI.

Looking forward to your reply. Thank you for accepting my join request.

Best Regards,
Arushi

--
You received this message because you are subscribed to the Google Groups "vocbench-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vocbench-use...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/vocbench-user/2c0f88ba-8420-4fa0-bc93-f511142e7da4n%40googlegroups.com.


--

Thomas Francart - SPARNA
linked data | domain ontologies | knowledge graphs
blog :
blog.sparna.fr, site : sparna.fr, linkedin : fr.linkedin.com/in/thomasfrancart
tel : 
 +33 (0)6.71.11.25.97

stel...@uniroma2.it

unread,
Mar 20, 2026, 9:50:45 AM (21 hours ago) Mar 20
to Thomas Francart, Arushi T, vocbench-user

Oh, thanks Thomas for picking this up! I was sending my reply and saw yours coming at the same time :-)

 

Well, the general answer for sheet2rdf and for the two other solutions I gave in my previous email is that yes, if it is possible through the UI it’s possible to invoke related APIs. In the specific case of Sheet2RDF my recommendation would be to use the wizard in the UI to prepare the general conversion once an Excel format for the mind map tool has been provided. This way, a PEARL file for the conversion will be generated and then it will suffice to use only the API, passing the PEARL rules.

 

We do not publish a complete list of Web API (there are hundreds of API calls), but we prepared a guide that instructs on different possible ways to learn the APIs to use:

 

https://semanticturkey.uniroma2.it/doc/user/web_api.jsf

 

Kind Regards,

 

Armando

 

Arushi T

unread,
Mar 20, 2026, 10:00:34 AM (21 hours ago) Mar 20
to vocbench-user
We had tried Sheet2RDF, but we kept getting stuck at the PERL mapping stage. It is possible that we missed certain steps to correctly prepare the Excel file to invoke the PERL mapping. The mind mapping tool we are using is called MindMeister. We are using it in the free version, and the export format is a .mind file, which is an archive file with a JSON file inside. The JSON objects already have some identifiers for each concept, but these do not align with the UUID URI generator layer of VocBench. Therefore, we would like to import into VocBench this time with all concepts following the same URI generation logic.

stel...@uniroma2.it

unread,
Mar 20, 2026, 4:34:01 PM (14 hours ago) Mar 20
to Arushi T, vocbench-user

Dear Arushi,

 

for the JSON file, there is no direct adaptor (and obviously there couldn’t be one a-priori, as JSON is just a syntax, then the organization of knowledge varies and has no direct mapping to a specific RDF vocabulary). So, you still have the three ways I mentioned earlier, more specifically and tight to your case:

 

  1. Write a simple script that reads the mindmeister JSON file and invokes API for creating concepts

 

  1. Use Sheet2RDF’s UI for creating the mapping and testing it the first time. Then you can use API as I mentioned in the previous email. For how to use Sheet2RDF, I linked the online pages on the VB site, plus you have the very good guide written by Thomas. Start with simple data: the concept and a few more stuff, then you can improve progressively. And you don’t need to write PEARL directly, the wizard can guide you through it. And if you are stuck at a basic point, you can post the Excel here, asking for help (pls keep it general, this group is for sharing information that are useful to all of its members)

 

  1. Develop a lifter plugin for VB, specifically created for the mindmeister export format. You don’t need here to use WebAPI, rather just the Java API

 

If the output from MindMeister (I used it many years ago, so I don’t have a clear memory of it nor I’m updated on its latest features)  does not contain much information about the concepts (e.g. just 1-3 different properties for linking concepts to other concepts) then I think option 1 is the best one

Conversely, in case you have lot of arcs in your graph described in MindMeister, a mid-step through Excel is worth the work

I’d recommend 3 only if you are consistently and regularly using it on the long run

 

Kind Regards,

 

Armando

 

Reply all
Reply to author
Forward
0 new messages