Survey Of Editing Tools For Building Ontologies

Michael Denny writes "Ontology Building: A Survey of Editing Tools is an up-to-date summary of more than 50 software tools for creating and editing ontologies. A brief introduction to the nature of ontologies and ontology building is included."

Comments

WTF is an Ontology?

[Anonymous Coward]

Nobody is participating in this discussion because we haven't a clue what "Ontologies" are...

Blah blah blah... this sentence does not make any sense to non-ontological people...

The semantic structuring achieved by ontologies differs from the superficial composition and formatting of information (as data) afforded by relational and XML databases. With databases virtually all of the semantic content has to be captured in the application logic. Ontologies, however, are often able to provide an objective specification of domain information by representing a consensual agreement on the concepts and relations characterizing the way knowledge in that domain is expressed. This specification can be the first step in building semantically-aware information systems to support diverse enterprise, government, and personal activities.

Re:WTF is an Ontology?

[HughsOnFirst]

At first blush, I'd guess that "ontology" editors would change my "ontology".
Something like a concussion , a nasty hangover , hallucinogenic drugs or maybe a graduate degree?
Oh wait, maybe this is for editing other peoples "ontologies" ? Maybe some sort of religious war is involved?

But seriously, This looks like good stuff with a terrible name,

Computer Engineering == Economics?

[Quixotic Raindrop]

[rant mode=on]
I have long despised the "science" of economics because they have an annoying tendency to take common, everyday words, with well-defined meanings, and turn them into something completely unrelated (see: efficient). Now, computer science and knowledge engineering is doing the same thing? Ontology already has a specific, well-defined meaning, and it has nothing whatsoever to do with the scope of agent or community relationships and concepts.

When describing a concept that does not yet have a descriptive word or phrase, don't just assume that you can take a word out of the dictionary and co-opt it for your own use. English is a hard enough language without our academics and researchers stealing words and twisting them to new, completely unrelated topics. If no word fits, make one up!!
[/rant]

Re:Computer Engineering == Economics?

[Jon Peterson]

Hey, come on, we can live with a little operator overloading in natural languages.

The only other time I've come across ontology is in philosophy, where it has a precise and well understood (among philosophers) meaning. I don't think there will be much general confusion.

I agree that the sort of behaviour you talk about is bad when it causes an old meaning of a word to become useless, because people confuse it with the new meaning. In this case we are talking about a rather arcane word being co-opted from philosophers by information scientists. There's little reason to use the word in general conversation, and little chance that either of the two groups above will confuse the word or use it sloppily.

Now, the abuse of the word 'enormity' is a different story altogether...

Re:Computer Engineering == Economics?

[fuzzbrain]

I think you're being a bit hard on these people. In philosophy ontology is the study of what exists. With knowledge engineering the question is more like what SHOULD exist. But when you look at philosophical arguments, you see that the questions really collapse into one another, and the philosophical question becomes one of what should exist in order for us to make sense of the world: forms, ideas, windowless monads, or just facts. In the context of a semantic web, I imagine that similar arguments will need to be made as to what the building blocks of that wordl will need to be. I think the use of 'ontology' in this context is both useful and appropriate.

Re:Computer Engineering == Economics?

[p3d0]

Computer science (and mathematics) has a long history of redefining common words. You have "assemble" and "compile", "class" and "type" and "method" and "code", "file" and "string", "stack" and "heap" and "hash", and so on. None of these has their original english meaning when used in computer science.

Little help people

[Henry V .009]

Anybody know how to turn the volume down on a bullshit dectector? My eardrums are bleeding, I swear.

I have priorities

[L. VeGas]

I don't so much want to learn about ontologies. I want to learn what an ontology is.

Re:I have priorities

[sql*kitten]

I don't so much want to learn about ontologies. I want to learn what an ontology is.

It's a way to leverage the synergy of paradigms.

Or in otherwords, the CS equivalent of management consultant bullshit.

This is what happens

[nosferatu-man]

... when you let people with grant money and "big ideas" near a concept that they don't understand. They hijack it and produce mountains of meaningless buzzword babble, trying to puff up their own particular snake-oil prescription.

"Ontologies" indeed. I bet David Hume would loooooove this.

'jfb

Ontology:

[charlie763]

"Webster's Revised Unabridged Dictionary (1913)"
Ontology On*tol"o*gy, n. Gr. ? the things which exist
(pl.neut. of ?, ?, being, p. pr. of ? to be) + -logy: cf.F.
ontologie.
That department of the science of metaphysics which
investigates and explains the nature and essential properties
and relations of all beings, as such, or the principles and
causes of being.


I still don't get it...

Can't...

[CableModemSniper]

Think...Buzz...word...Overload...

Heh

[4of12]

At least they're developing something useful!

For a while, I was afraid they were developing ways for studying tautology!

Simple Explanation

[moc.tfosorcimgllib]

A simple computer ontology builds abstract relationships between objects.

Animals -> Birds -> Flightless Birds -> Penguins
It is an important first step into AI and computer interaction.
Unfortunately I found the article as useful as those old programs where the computer would try and guess what you were thinking about.

Re:Simple Explanation

[Anonymous Coward]

That's interesting.

Tell me more about this.

Re:Simple Explanation

[Anonymous Coward]

That's interesting. Tell me more about this.

Damn! Elisa is trolling again.

Re:Simple Explanation

[Tablizer]

A simple computer ontology builds abstract relationships between objects. Animals -> Birds -> Flightless Birds -> Penguins

Are they always hierarchical? I think hierarchies are overused, or at least oversold, in computer science. I have kicked around some set-based classification systems and interfaces here. It includes an example college store which needs to classify its products. Hierarchies don't work well because a given item can fall under multiple categories.

The Dewey Decimal System sometimes has this problem also. For example, does a book about computerized gene studies go under the Biology section or the Computer section? Ideally, it should go under both. Sets allow this, while nested classifications don't (without node duplication).

Re:Simple Explanation

[Anonymous Coward]

The main use of ontologies (as I recall) is for classification of documents. The idea is to take a big pile of documents, run it through an "engine" and out pops a searchable system that has classified each document according to its content. Presumably, this would be like generating a Yahoo directory automatically. For example, it could determine that since a document contains references to a known company, that dollar amounts are presented, and that certain keywords are present thus the document is an annual report!
I believe that Interwoven (www.interwoven.com) has this kind of technology. Probably called meta data extractor or something.
The document and content management folks love this kind of stuff. So check out the main vendors like Documentum, Interwoven, and Docushare.

Re:Simple Explanation

[Fizyx]

Animals -> Birds -> Flightless Birds -> Penguins

Yes, that's how it is being used. That's NOT what ontology means.

Taxonomy would be a better word for this.

Re:Simple Explanation

[rpg25]

Actually, the advertised purpose of the ontology projects was to permit interchange of information between different intelligent systems. Ideally, the definitional information contained in the ontology was to permit different reasoning systems to understand each other's representations and communications. I would say the jury is still out on the utility of this approach.

There have long been systems that attempted to describe objects and their relationships (all the way back to Minsky's frames in the late 60s or early 70s. The ontologies were supposed to have some semantic groundwork that would let them be interchangeable and allow translation between different ontologies.

Similarly, on the web, ontologies for web page contents were intended to permit programs to grok the meaning of those web pages.

Unfortunately, a lot of effort seems to have gone into syntax and semantics of the ontologies themselves. This leaves some holes in:

1. The process of translating between ontologies
2. Figuring out where all that semantically marked-up content is going to come from.

RTFA

[panurge]

OK, the article is a bit vague, because it is aimed at people who kind of know what an ontology is already, and it is really just an overview of the current state of the art.

However, I found it useful if only for one part, where they talk about ontology construction. If you read it, it looks just like a systems analysis job slightly rephrased. Find out what the entities are, establish the relationships between entities, logicalise, rationalise, and finally populate the resulting structures. There, that wasn't so hard, was it?

It looks to my simple and fast ageing mind as if we end up with something like a DFD in which data stores are replaced by sets and data flows are replaced by relationships, and I have no idea at all what happens with processes. Having done a bit of KB work in the late 80s and then failed totally to keep up with the field, I'd like to know more at a practical level, but without having to understand medical applications. Anybody got any good links?

Practical useability for slashdotters

[beofli]

In our information jobs, we deal with different kinds of knowledge: code, documents, websites, personal information. For each type tools exists to browse, edit and query, but when it comes to information that overlaps different types of knowledge, we are stuck with manual labour and guessing.

Example: Who made requirements document X and which code implements this?

In the (I hope near) future, Operating systems are replaced by ontology editors, where instead of "files" we work with a network of accociated data, where ourselves, and agents, will navigate through, in a intuitive manner.

Re:Practical useability for slashdotters

[Anonymous Coward]

Yes. That OS is called Lisp...

What is an Ontology? Explained here.

[Anonymous Coward]

See this web page for what these buzzword-merchants actually mean.


Short answer:
An ontology is a specification of a conceptualization.
The word "ontology" seems to generate a lot of controversy in discussions about AI. It has a long history in philosophy, in which it refers to the subject of existence. It is also often confused with epistemology, which is about knowledge and knowing.

In the context of knowledge sharing, I use the term ontology to mean a specification of a conceptualization. That is, an ontology is a description (like a formal specification of a program) of the concepts and relationships that can exist for an agent or a community of agents. This definition is consistent with the usage of ontology as set-of-concept-definitions, but more general. And it is certainly a different sense of the word than its use in philosophy.

What is important is what an ontology is for. My colleagues and I have been designing ontologies for the purpose of enabling knowledge sharing and reuse. In that context, an ontology is a specification used for making ontological commitments. The formal definition of ontological commitment is given below. For pragmetic reasons, we choose to write an ontology as a set of definitions of formal vocabulary. Although this isn't the only way to specify a conceptualization, it has some nice properties for knowledge sharing among AI software (e.g., semantics independent of reader and context). Practically, an ontological commitment is an agreement to use a vocabulary (i.e., ask queries and make assertions) in a way that is consistent (but not complete) with respect to the theory specified by an ontology. We build agents that commit to ontologies. We design ontologies so we can share knowledge with and among these agents.


You too can do in XML+Java, and at great investment of time, what is already possible with Lisp. But hey, it looks like you're doing work and you might get some venture capital. Or not, in the current market...

Re:What is an Ontology? Explained here.

[Anonymous Coward]

You'd have to do lots of work in Lisp too (but it WOULD be easier - still lots of work though.).

Ontology recapitulates phylogony..

[Anonymous Coward]

or something like that. But what does Biology have to do with /.?

Missing information

[webmaven]

For some reason, the table doesn't list the license or price of any of the surveyed tools. Only one tool even has the phrase 'open-source' in it's 'Notes' column. One other says 'source available', and a third seems to be hosted on sourceforge.net. Should I simply assume that the rest of these tools are proprietary?

Given the fact that I am (on occasion) willing to part with my $$$ for software if necessary, I definitely would have liked to see pricing information included in this table, if only to rule out those tools which are out of my price-range.

Very Interesting Piece...

[Master of Transhuman]

Has to do with an aspect of conceptual processing, an AI area which is the core of AI and gets very little work done in it...

But without it, you'll never have AI - nor will you ever have intelligent systems development technologies - which means the crap being sold today will continue to be sold...

So get on it...

A website about Ontology

[exa]

I'd made a research on Ontology Tools for Repositories on Internet including a quite satisfactory survey of ontology subject and a preliminary ontology language design for a grad. course given by Varol Akman at Bilkent CS. Dept.

It answers the quintessential question of "What is an Ontology?"

Thanks,