Need Feedback Reviewing Journal Submission Neo4J/OWL? Patents and Code

[Michael DeBellis]

I'm reviewing a paper for an ontology journal that seems somewhat confusing. They talk about OWL but they use Neo4J which isn't natively OWL or RDF compliant. I looked on the Neo4J site and they have an article about how you can import an OWL ontology into Neo4J but although it isn't clearly stated, my strong feeling is once you do that you are essentially working in the Neo4J model from that point on. For example, you couldn't run a reasoner on the ontology after you load it into Neo4J. The authors of the paper don't seem to know this, at least it isn't clear that they do and they talk about Neo4J and OWL almost as if they are interchangeable. But before I give them feedback I wanted to double check because who knows what kind of progress Neo4J might have made. Does anyone know is it possible to still treat a model as OWL after you load it into Neo4J? E.g., can you run any OWL reasoners on an OWL model that has been imported into Neo4J? I'm almost certain the answer is no, for one thing the paradigms aren't consistent. Neo4J supports essentially ternary relations (like RDF*) because it explicitly represents properties as objects which themselves can have properties and OWL only supports binary relations because more than that would be too hard for the reasoner. 

On another issue: they say that they haven't released their code on Github because it is pending a patent. First, from the quality of the paper I'm amazed that someone is trying to patent it... although I've also had some experience with patents and the people who grant them often aren't the most technical and I've seen patents granted on some ridiculous ideas.  (E.g., I'm pretty sure someone patented the idea of a shopping cart) The issue is when you go to enforce such patents I think the bar gets higher because then you are dealing with people who know their stuff, not patent bureaucrats. Anyway, I'm curious what people think, my bias is if you write a paper about some code that you developed you need to make the code available or I don't believe you. Does patent pending sound like a legitimate excuse? Is it consistent with publishing in a journal which is meant to share your work with the community at large to then say you can't share your code because it is pending a patent? 

I want to be fair and would welcome any feedback.

Michael

[Chris Mungall]

Hi Michael

I assume you have seen some of the work of Jesús Barrasa, who is behind neosemantics (n10s) (https://neo4j.com/labs/neosemantics/)

This describes some basic ABox-type inferencing that can be done with a plugin:

https://jbarrasa.com/2019/11/25/quickgraph9-the-fashion-knowledge-graph-inferencing-with-ontologies-in-neo4j/ 

But you are right that OWL semantics is defined in terms of ABoxes of triples. I think there is an opportunity for the KR community to provide a native semantics for property graphs. There is a lot that gets discussed on this forum that is relevant. E.g. IKL's quoted sentences ("that") maps into how a lot of people use edge properties in graph databases like Neo4J, and this is something that is slowly getting into the conventional semweb community via rdfstar.

See for example:

https://github.com/linkml/owlstar

--
All contributions to this forum are covered by an open-source license.
For information about the wiki, the license, and how to subscribe or
unsubscribe to the forum, see http://ontologforum.org/info/
---
You received this message because you are subscribed to the Google Groups "ontolog-forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/45d73a04-2add-4ec5-a70f-c47c239c16b2n%40googlegroups.com.

[Alican Tüzün]

Hello Michael,

I played with the neoj and as Chris mentioned, there is a plugin called neosemantics, which lets you do the mapping between the data and the ontology resulting in a "knowledge graph", however complex reasoning  was not possible.

Also not sure if any reasoner can solve complex structure emerged from property graphs.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAN9AifuVLVyg%3D-oPVc9EVfjaFcnkEG6O6DvLXfoK%3DtfNsUa6hw%40mail.gmail.com.

[Michael DeBellis]

Thanks, that's what I thought but wanted to be sure. 

Michael

You received this message because you are subscribed to a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/n7-naYmLncc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHzn%3DSA3cV_9OMZAXSdP%2BDHgg%2B%2BfPdBU_y3MXNUQ6cSojQ%40mail.gmail.com.