From: Atilla Elçi (DAÜ) [atilla.elci@emu.edu.tr]
Sent: Perşembe 13 Ocak 2005 11:06
To: 'CmpE583Fall2004'
Subject: [CmpE583Fall2004] Remarks on Draft Reports.

Importance: High


Good morning all,

Thank you for turning in your draft reports; I found them all well done,
however, requiring considerably more development. I've reviewed all draft
reports and put my remarks on them. You can collect your reports from my
office.

For the benefit of all, and to further stress requirements, here I summarize
my findings and remarks below:

1. Importance of report format cannot be over emphasized. Please use the
format provided in 401 site. This calls for coverage of introduction,
development, results, discussion, conclusion, properly referenced, and major
code/data/interfaces in appendi(-x/ces). 

2. Furthermore, as it is not clearly stated there, you should also include a
subsection on the platform (the case) utilized. This means you will
introduce and explain the case as to its significance, technology base,
parts and operation.

3. Furthermore, a proper coverage of the case study, that is, your scenario,
must be included. This should involve your hypothesis, study done to show
that, samples runs and results obtained, discussion of 'why's and 'why not's
leading the reader to your conclusions. These all go in their proper
sections of the report: hypothesis usually takes place in introduction and
may be repeated in main to further clarify it; study and results take place
in main development; and the rest in conclusion.

4.  All of you develop some ontology for your case study; and most of you
use Protégé and supply your ontology as directory. This is not wholly
acceptable. The most inportant part of your case study is probably your
ontology and you should dwelve on it at length. This may be through
introduction of clusters and explanation of why its the way it is. You
should try represent your ontology in various other forms in the way of
presentation, such as UML class diagram, EER diagram, ontology graph,
OWL/RDF/XML, etc.

5. As always has been my request from you, originality of your report is
essential. It should be composed in your own words. Any quotes should be
properly identified so; all important concepts must be properly introduced,
no matter shortly, and referenced. 

6. All not-so-essential and lengthy material, that is, more than 1/2 a page,
such as code, graphs, and excerpts should be placed in the appendices.
Others may be interspersed between your writing.

7. No one tried to validate his/her ontology but relied on Protégé instead.
Operationally this is OK. However, in a case study such as yours, you should
strive to test your ontology before using it. Try find an ontology
validator; if no general validator is available, try an OWL validator/parser
in order to verify your work.

8. One last item of importance is the very fact that this exercise is on web
semantics. Your case study should strive to show use and utility of SW
approach. This involves a judicious choice of search queries if one is
involved or properly conditioning of your scenario and hypothesis. Of course
you'll use relational support but that should not be all there is to your
case study. A proper study in this respect, compares and contrasts, where
appropriate, SW approach against other approaches, primarily against
relational model. So, try same queries in SQL for example and discuss the
results.

9. One absolutely-the-last item concerns the English of your reports. Please
take interest, have some one you trust review it, and try do through some
means better than your best to improve your reports presentation.

Warm greetings to you all.

Yeni yılınız kutlu olsun. /Wishing you a happy new year.
Atilla Elçi
http://cmpe.emu.edu.tr/aelci/ 




 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/CmpE583Fall2004/

<*> To unsubscribe from this group, send an email to:
    CmpE583Fall2004-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/