consolidation of definitions
Alex Shkotin
Hi John Antill and All,
The current versions of pages https://ontologforum.com/index.php/Ontology(IT) and https://ontologforum.com/index.php/Knowledge_base represent John Antill's work well. Thanks to him for that. However, according to rule "(Roaod) one author one definition", a collection can only have one current definition from a single person. Therefore, consolidation is necessary.
I asked chatGPT to consolidate the definition. The results are in [1] and [2].
Accordingly, only the consolidated definition will be posted on the page.
John can confirm or correct the chatGPT consolidation.
Please keep in mind that I have not verified the AI's reasoning. Only the author can provide convincing verification.
So far, candidates to remain on the pages from John Antill: JA004, JA001N.
John, your turn 🤝
Alex
[1] !IT-ontology
I'll give you several definitions of the term "IT-ontology." Your task is to formulate a consolidated definition from which they all follow. However, if any of these definitions proves to be the most comprehensive, just take it.
JA001!
Ontology – formal semantic model enabling logical inference.
JA002!
An ontology in IT is a formal artifact with explicit semantics.
JA003!
A formally specified conceptualization expressed in a representation language with explicit semantics.
JA004!
A knowledge base whose artifacts constitute explicit axioms about domain entities and relations, expressed in a language with formal declarative semantics.
JA005!
IT ontology is a formally encoded artifact with defined semantics.
Among the given options, JA004 is already the most comprehensive: it includes the ideas of formal artifact, explicit semantics, conceptualization of a domain, and support for inference through axioms. The other definitions can be derived from it by simplification.
A slightly consolidated formulation that preserves all elements present in the set would be:
Consolidated definition
An IT ontology is a formally encoded knowledge artifact that specifies a conceptualization of a domain as explicit axioms about entities and relations, expressed in a representation language with formal declarative semantics enabling logical inference.
Why the other definitions follow
JA001: If an ontology is expressed with formal declarative semantics and axioms, it functions as a formal semantic model enabling logical inference.
JA002: The consolidated definition explicitly states it is a formal artifact with explicit semantics.
JA003: “Specifies a conceptualization of a domain expressed in a representation language with explicit semantics” follows directly.
JA004: Essentially identical, only slightly expanded into a general definition.
JA005: A shortened version of the same idea: formally encoded artifact with defined semantics.
If desired, I can also produce a minimal canonical definition (ISO-style, ~20 words) from this consolidated one.
[2] !KG
I'll give you several definitions of the term "KB." Your task is to formulate a consolidated definition from which they all follow. However, if any of these definitions proves to be the most comprehensive, just take it.
JA001!
A knowledge base is any structured or semi-structured repository of information designed for retrieval or inference.
JA001N!
A knowledge base is a structured or semi-structured repository of information intended for systematic retrieval or inferential use.
JA002!
A knowledge base is a structured set of knowledge artifacts that supports inference.
JA002N!
A structured set of knowledge artifacts in a representation supporting inferential use.
Among the four definitions, JA001N is the most comprehensive. It contains all the elements that appear in the others and can generate them as special cases.
Why JA001N is the most general
JA001N:
A knowledge base is a structured or semi-structured repository of information intended for systematic retrieval or inferential use.
It includes the following components:
Repository of information → this can be interpreted as a set of knowledge artifacts (JA002, JA002N).
Structured or semi-structured → covers both stricter representations (as in JA002N) and looser ones.
Intended for systematic retrieval or inferential use → includes:
retrieval (JA001)
inference (JA002, JA002N)
How the other definitions follow
JA001
Remove “systematic” and keep the same structure:
repository of information designed for retrieval or inference.JA002
Interpret repository of information as a set of knowledge artifacts, restrict the purpose to inference, and require structure.JA002N
Further specialize:repository → set
information → knowledge artifacts
structured/semi-structured → representation supporting inference
Result
The consolidated definition is therefore simply:
A knowledge base is a structured or semi-structured repository of information intended for systematic retrieval or inferential use.
All other definitions can be obtained from it by restriction of terms or scope.
alex.shkotin
Consolidation also requires two definitions from BL.
And again, we'll use AI. Especially since JA agreed to the AI's work and its definitions.
IN:
I'll give you two definitions of the term "KB." Your task is to formulate a consolidated definition from which they all follow. However, if any of these definitions proves to be the most comprehensive, just take it.
BL001G!
A knowledge base is a centralized repository of information.
BL002!
A knowledge base is a machine-readable resource for the dissemination of information.
OUT:https://chatgpt.com/share/69cead05-d3c8-832d-9512-bb0702ac7446
A minimal consolidated definition that entails both of yours would be:
A knowledge base (KB) is a centralized, machine-readable repository of information designed for storage and dissemination.
Why this works (briefly):
“repository of information” → covers BL001
“centralized” → preserves BL001 explicitly
“machine-readable” → captures BL002
“designed for dissemination” → captures the functional aspect of BL002
If you want it even more abstract (closer to a theory/model distinction you like), we could generalize further:
A KB is a structured information resource that supports representation and transmission across agents (including machines).
Say if you want a stricter formalization (e.g., in terms of models, languages, and interpretations).
alex.shkotin
These two BL's definitions are a good example of inline definitions. This is when the text cited as a knowledge unit can be found inline in the source text.
In the citation from the page "What is a Knowledge Base? Definition, Examples, and Guide," the text of the two inline definitions is underlined:
"What is a knowledge base?
In general, a knowledge base is a centralized repository of information. A public library, a database of related information about a particular subject and whatis.com could all be considered examples of knowledge bases.
In relation to IT, a knowledge base is a machine-readable resource for the dissemination of information, generally online or with the capacity to be put online. Knowledge bases are an integral component of knowledge management systems. They are used to optimize information collection and information organization and retrieval.
A digital knowledge base is not a static collection of information but a dynamic resource. They may themselves have the capacity to learn, as part of an automation or artificial intelligence expert system.
"
Well, the consolidated definition is now in our collection.
tkn...@neurocollective.com
--
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 visit https://groups.google.com/d/msgid/ontolog-forum/e5a79605-c5f1-409d-965a-e6f22b2247d6n%40googlegroups.com.
Nadin, Mihai
Sent: Friday, April 03, 2026 11:49
To: ontolog-forum <ontolo...@googlegroups.com>
Cc: ontolog-forum <ontolo...@googlegroups.com>
Subject: Re: [ontolog-forum] Re: consolidation of definitions
Alex Shkotin
Tom,
Since we're compiling a collection, it's never too late to add. If you'd like, I'll add your definition as a unit of knowledge as "A knowledge base is a centralized repository of curated knowledge - including facts, relationships, rules, and context - organized to support efficient retrieval, understanding, and reasoning." or whatever you'd like.
What I need is your willingness to contribute to the definition collection.
Discussing the relationship between different definitions is a separate, delicate matter for the Discussions section.
As for the terms Information(IT) and Data(IT), that needs to be discussed separately. I'm intuitively inclined to the idea that this is a primary term in Information Theory that won't have a definition, but will instead have an axiomatic definition of its relationship with other primary terms through primary relations.
For me, an example of such a definition, as applied to geometry, is Hilbert's axiomatic theory Foundations of Geometry.
Meanwhile, in modern deterministic IT, information and data are located somewhere in the computer, which is a processor and finite memory. When we fill memory and start the processor, it begins to change the memory (roughly as Turing or Post described), and we create a theory of how this particular processor moves through this particular filled memory. And at some point, we might need the term "information."
We can start with the fact that any state of memory is filled with data.
Some of this data are instructions for one processor, but not for another.
A most interesting topic.
So, would you like to add your definition of KB to our collection?
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/206994525576491008.0.v2%40titan.email.
Alex Shkotin
Tom and All,
One of the remarkable situations we encounter when comparing different definitions of the same term in the same field (IT): we need to conduct some kind of reasoning to prove or disprove that the definitions define the same class of objects or processes. Like the well-known definitions of a triangle: one has three angles, and the other three sides.
And such reasoning is worth exploring separately. Moreover, it seems to me that when we examine the rules applied at any given step of the reasoning, they will be far removed from abduction, deduction, induction, modus ponens, etc.
We already have two examples of such reasoning from chatGPT.
John F Sowa
Sent: 4/4/26 8:20 AM
Alex Shkotin
John,
I think the idea of collecting definitions of key terms in our field and discussing their consolidation would be useful.
I haven't seen any significant activity yet. Even Tom hasn't responded yet about adding his definition.
John Antill found many interesting definitions.
Tom Knorr offered his own.
So far, definitions of two terms have been collected https://ontologforum.com/index.php/Category:Definition
The next term hasn't been decided yet.
Let's discuss any details of this project, which is primarily a collection.
And that's a special kind of enthusiasm.
For now, please take a look at this document:
https://ontologforum.com/index.php/%22DEFINITIONs._rules_and_tasks%22
How can it be improved?
--
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 visit https://groups.google.com/d/msgid/ontolog-forum/3f82d810008a4cd1be63722b9c625401%4094af56dc829e416bad7ccc8bd33622b0.
tkn...@neurocollective.com
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORBuCQAOqdhAt0VJVEe%2B6NiLY%3D%2BJr4qMNKYPqu30Nu17w%40mail.gmail.com.
Alex Shkotin
Tom,
We're just developing the procedure for adding a new definition, and right now all we need is an author and a desire to do so.
Although at some point we'll probably add a request for the author to read existing definitions and state their disagreement with them.
After all, most people will probably find something that aligns with their opinion.
And any analysis of the collection will be in the Discussion section, or will link to the collection.
I'll add it tomorrow.
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/207148658410385408.0.v2%40titan.email.
Alex Shkotin
Alex,
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/207148658410385408.0.v2%40titan.email.