root analysis questions
Raheel Ashraf
Mohsen Saboorian
On Sun, Nov 23, 2008 at 6:27 PM, Raheel Ashraf <mrad...@yahoo.com> wrote:
> AsSalamAlykum
> jazakallah khair for your earlier reply. I am very interested in the
> analytical aspects of the quran and hence the root word analysis. Here is
> what i need to find out:
> - Can you please give me the contact infor for someone working at Computer
> Research Center of Islamic Sciences (CRCIS). I need to find out if they have
> certain data that i am looking for
Sure. Here is the website:
http://farsi.noorsoft.org/index.php?newlang=english (the URL is a bit
contradictory ;)), and this is the email address to contact:
in...@noorsoft.org
> - The root search feature, does it list all of the root letters and their
> combination that occur in the quran.
Yes it does.
> - Can you point me to the places in the code where i should start looking at
> to find the root search feature implemented and other things related to it
Please refer to classes under package: net.sf.zekr.engine.root.
Mohsen.
Raheel Ashraf
From: Mohsen Saboorian <moh...@gmail.com>
To: ze...@googlegroups.com
Sent: Sunday, November 23, 2008 1:43:27 PM
Subject: [zekr] Re: root analysis questions
Mohsen Saboorian
On Tue, Nov 25, 2008 at 7:45 AM, Raheel Ashraf <mrad...@yahoo.com> wrote:
> AsSalamAlykum
> I was thinking of adding the reverse root analysis feature. Subhanallah, i
> looked at the code and saw that you had the same feature in mind for the
> next release. Mashallah. I saw some code already in QuranRoot. Can i work on
> this feature? Are there other places where you also put commented code to
> support this feature? Whats the UI you have in mind for this feature
The commented code in QuranRoot works fine, and as far as I remember
it's just there that should be uncommented. I commented it out for now
because it adds an average loading time overhead of 1s.
Unfortunately the mapping between root and Quran locations is not
currently one-to-one in the root db. For example we have two roots: 1)
محمد and 2) حمد. These cases can be easily detected by an smart
processing when adding reverse mappings (which is not currently
coded):
See as an example the above words: 1 is a subset of 2. So we have more
(or equal) occurrences of 2 than 1. I didn't verified it yet, but I
think this can be resolve by means of comparing the number of
occurrences for roots whose mappings conflict.
Regarding the UI, here is what I had in my mind:
User selects "Analysis mode" (two-state menu item) from a menu. This
causes a JavaScript code to be executed on the page which then makes
all the Arabic words a hyperlink. When user clicks on a link, it
brings up a small window in which user can see the root. This window
can later show more info of a single term.
Another (more easier) scenario is:
Press "Analyze" from View menu, and it brings up a new window in which
there is a table of two columns. In the first column the words are put
one by one. In the second column, corresponding root can be found.
> Secondly, I think we should look at verifying the list of root words from
> other sources too. I know there are other websites that have similar listing
> so maybe we cna write some comparison code that compares this list to others
> inshallah...
I appreciate if you can compare it with another root database.
Regards,
Mohsen.