- From: Peter Brooks <peter.h.m.brooks@gmail.com>
- Date: Sat, 13 Jul 2013 10:18:29 +0200
- To: Cristiano Longo <longo@dmi.unict.it>
- Cc: semantic-web@w3.org, s.rizza@gmail.com
It depends on the sort of questions you expect to be answering and who is going to be using the ontology. You might be able to define an ontology that can give you logical connections between restaurants that serve pizza and pizza restaurants, but that's unlikely to be very useful to anybody and, besides, the first is a superset of the second. To be useful you certainly need to have a structure that allows logical queries, but that's simply a given, rather than what you need to establish a sensible and useful ontology. Vocabularies for menus are widely available with simple google searches. Here are two examples: https://www.mrd.com/browse http://www.constantia-uitsig.com/media/download_gallery/CU%20MENU.pdf I think that a useful ontology would allow somebody to create a query that would return results from both sort of establishment and allow comparisons to make decisions on the best one to use for a particular requirement. On 13 July 2013 10:06, Cristiano Longo <longo@dmi.unict.it> wrote: > I adopt a formal approach based on first order logics and unit testing for > modelling (see for example > http://webont.org/owled/2013/papers/owled2013_15.pdf). However, I'll > consider advantages and fallbacks of your proposal in these days. I was just > asking about vocabularies for the knowledge domain of restaurants and food. > > CL > > > On 13/07/2013 09:43, Peter Brooks wrote: >> >> I like this idea because it has a well-defined and, I hope, fairly >> narrow scope. So it might be excellent for educational purposes in >> ontology as well as actually useful for restaurant comparisons and >> reviews one day. >> >> I'll be embarking on an architecture and ontology exercise very soon >> (the details are here if anybody is interested: http://ow.ly/mTZex ), >> so this might be a good sample of the sort of development process >> required to help those attending who are not so familiar with >> ontology. >> >> My intention was to use the 'dot' language (grapviz) as the >> development tool during the fact-to-face sessions because it is really >> quick and easy to work with as well as giving good, quick, graphical >> representation. This should help, too, in decomposing the top-level >> architecture into digestible chunks. >> >> These 'dot' files can then be translated to JSON and, if necessary, to >> OWL. >> >> My intention, after that, was to define Ada packages for the >> architecture and ontology to enable ease of use by those interested in >> reliable software infrastructure and to enable an enterprise service >> bus (like TIECO / SOA). >> >> Would that approach interest you? >> >> On 13 July 2013 09:25, Cristiano Longo <longo@dmi.unict.it> wrote: >>> >>> Hi, I'm approaching modelling and publishing information about menus of >>> restaurants. Some suggestion and/or collaboration proposal? >>> >>> Thanks in advance, >>> CL >>> >> >> > -- Peter Brooks, Fellow in Service Management (FSM)® Tel: +27 21 447 9752 Mobile: +27 82 717 6 404
Received on Saturday, 13 July 2013 12:07:10 UTC