Tomorrow, the seminar-class has been canceled as our professor is attending a seminar of his own. It has been rescheduled to Thursday I’ll come back to that then.
While our professor is currently out of town, and even out of our country, we’ve stumbled upon several important questions regarding the transformation of XBRL into OWL, and thus ontologies.
Due to the fact an ontology is designed with several questions which cannot be answered easily in mind, and an XBRL-taxonomy is already set to answer some very specific questions (What was last year’s revenue? How much did we spend on housing? and so on), developing an ontology from that taxonomy is not of any use. When it is just to keep us busy, we can understand but while our professor isn’t able to explain to us what the added value of an ontology is during the group-sessions, we might as well ask him through an email.
And so it is. I’ve just sent an email to him posing several questions with the same subject: Why?
As we get to grips with our more existential questions, we also combined the already constructed ontologies and ‘instances’ to come to a better, more comprehensive and more elaborate version. We are very curious as to how our professor will grade our creation.
I’ll be posting updates soon!
3 thoughts on “Questions posed”
“EER: If so, I will proudly link to you :-)”
The time may be nigh, I think I will upload before the end of 2007 😛
That’s a pretty serious (and stimulating) question to ask a professor. I wonder if he (it’s Pijls, correct?) actually thought about that before creating the topic. If I recall correctly, this is the first time he treated this topic, as last year’s seminar topics were pretty straightforward (albeit fun, as far as I could tell) pathseeking algorithm thingamajigs (pun intended).
I think one thing you could take a look at is the structure of the questions that are meant to be answered with an ontology. For example, if an ontology is used to answer the question ‘Is this company capable of flexibly responding to an uncertain market event?’, you could use several pieces of information in an XBRL report to answer the question posed to the ontology, such as the standard ‘Percentage of revenue gained from new products’, which is generally used as an indicator of product flexibility in a company. Basically, it’s looking for cause-and-effect relationships. Of course, I’m not at all at home in Ontologies or XBRL, so I doubt I’m making any sense 😉
We should play some pool again soon =) then I can throw some more recruitment efforts in your direction 😉 I didn’t start about Business Intelligence yet – from these posts, that might be more in your direction than what I’ve been throwing at you so far =)
We’ll talk soon,
Timmey! Great to see you here, be sure to read here often as I will keep you all posted. (however strange that sentence might be.)
Yes, you are right about my professor.
And too: yes, that is what you want to achieve when using an ontology over straightforward XRBL. The cause-and-effect Ã¡nd the indirect links between several of the items within a company, like relations between total revenue and for instance housing costs.
Higher housing costs, often better places for your shops, higher exposure and thus higher revenue.
We most definitely have to play some pool again, I’ll get in touch with you in a short while. I’m also impressed by your factual knowledge, although it also might be because you’ve actually been reading all my previous posts… 😉