Spam detection software, running on the system "crl.nmsu.edu", has
identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email. If you have any questions, see
[email protected] for details.
Content preview: Thanks Sharaf for considering the comments and the feedback,
hopefully that will push the project forward. I would also suggest if
possible
-If you are not already working on it- to include additional weighting scheme
for the user to choose from, such as tf.idf or other summarization algorithms
(Maximum likelihood ..etc). This will provide different perspectives about
how to look at every chapter and also will allow to validate various
hypothesis
regarding surahs topics. [...]
Content analysis details: (6.1 points, 5.0 required)
pts rule name description
---- ---------------------- --------------------------------------------------
0.0 STOX_REPLY_TYPE STOX_REPLY_TYPE
2.7 TVD_FINGER_02 TVD_FINGER_02
3.4 FH_DATE_PAST_20XX The date is grossly in the future.
--- Begin Message ---
Thanks Sharaf for considering the comments and the feedback, hopefully that
will push the project forward.
I would also suggest if possible -If you are not already working on it- to
include additional weighting scheme for the user to choose from, such as
tf.idf or other summarization algorithms (Maximum likelihood ..etc). This
will provide different perspectives about how to look at every chapter and
also will allow to validate various hypothesis regarding surahs topics.
Regards,
Ahmed Abdelali
New Mexico State University
----- Original Message -----
From: "Abdulbaqi Sharaf" <[email protected]>
To: "El-Haj, Mahmoud" <[email protected]>; <[email protected]>
Sent: Monday, January 25, 2010 2:48 AM
Subject: RE: word cloud for Qur'anic chapters
Salam Bayan, Mahmoud, Eric and others,
thanks for your comments, this is just an initial attempt adopting this
open source script (http://www.lotsofcode.com/demo/tag-cloud-v2/)
As Mahmoud highlighted the core of the cloud is font size, color and
style, and I assure Bayan that the order is totally random, and you can
try the same sura twice and each occurance gives different order, I just
wanted to highlight more frequent words with bigger font, I hope I will be
able to enhance this with:
- richer set of function words
- better font style, size and color
- each word links to it's concordance and morphological feature from
Quranic Arabic Corpus
- including English terms as well
Mahmoud, my stoplist is just based on initial observation and still there
could be many more to include, and would appreciate using your list if
available as well as other resources you might already have.
thanks again for your comments...
best regards,
Abdul-Baquee M. Sharaf
PhD Student
Language Technologies Group
School of Computing
University of Leeds
UK
________________________________________
From: El-Haj, Mahmoud [[email protected]]
Sent: 25 January 2010 03:56
To: '[email protected]'; Abdulbaqi Sharaf
Subject: RE: word cloud for Qur'anic chapters
Dear Abdulbaqi,
Well done job and as Dr Atwel said it is about time for someone to do it
and I am sure it will lead to many interested and promising tools in the
future.
My previous work involved Qur'an Retrieving, we had a stop-words list in
addition to a light Arabic stemmer, and we also built a Qur'anic
thesaurus. I will be very keen to see their effects on the presented
Qur'anic Cloud if possible.
Regarding Bayan Shawar's answer, the most influencing parts in Clouds are
the font size or style, I do not think the order has anything to do with
it (for example, if we took more than one chapter “Sura” the order does
not make any sense as the repetition is eliminated). In my opinion, the
listing does not affect the Quran itself specially when it is unordered.
Best of luck,
Mahmoud
School of Computer Science and Electronic Engineering.
University of Essex.
Colchester CO4 3SQ, United Kingdom.
URL: http://privatewww.essex.ac.uk/~melhaj/
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Bayan Shawar
Sent: Sunday, January 24, 2010 8:16 PM
To: [email protected]; Abdulbaqi Sharaf
Subject: Re: word cloud for Qur'anic chapters
Salam Sharaf,
This is me Bayan Abu Shawar if you still remember. In fact it is a good
idea if you apply it in any Arabic corpus, but i am not sure what will be
the aim of doing this to Qur'an? Ok you extract the content words but they
are not appear in the same order as it in Qur'an, so it's like you extract
content words from different souras and merge it together. may be i am not
familiar with world cloud, but because it is Qur'an we should be careful.
If the aim is to extract th content word, then for example see how these
words are used nowadays in different Arabic corpra like media one,
comparing language of Qur'an with other ones, and do other useful
statistics, this will be good. I am just worry about the way result appear
in the link u sent us, just because it is qur'an, may be you could
presented in a different way. What do you think?
Good luck,
Bayan
--- On Sat, 23/1/10, Abdulbaqi Sharaf <[email protected]> wrote:
From: Abdulbaqi Sharaf <[email protected]>
Subject: word cloud for Qur'anic chapters
To: "[email protected]" <[email protected]>
Date: Saturday, 23 January, 2010, 16:29
Hello,
I have implemented "word cloud" for Qur'anic surahs:
http://www.textminingthequran.com/php/wordcloud.html
you can choose more than one Sura, and can consider only
content words, the list of stop words is initial and should
grow to more words in future, have a try and let me know
your feedback..
best,
Abdul-Baquee M. Sharaf
PhD Student
Language Technologies Group
School of Computing
University of Leeds
UK
--- End Message ---
