Salam Sumayya

I also found your project very interesting. It is something I am also
looking at.
For ready made themes of the Qur'an, I guess Nora's work is the most
complete data available.
However, if you consider linking verses together and find relationships
between them, and also bringing in Ahadith, then books of Tafsir (e.g., Ibn
Katheer) is good for that.

One tafsir I would recommend -if you or your team have command on Arabic- is
Adwa' al-bayan authored by Muhammad Amin al-Shinquity, which explicitly
establishes links between verses and completes the picuture where one verse
comes to add info not mentioned in another verse.

I have once compiled a set of over 40 examples from this book just to give
you a flavor:
http://www.textminingthequran.com/wiki/Understand_Quran_with_Quran

Like Kais, I am also curious to know about your architecture and data model
for this project.
best regards,

Abdul-Baquee
Leeds, UK

On Sun, Mar 7, 2010 at 12:33 PM, Kais Dukes <dukes.k...@googlemail.com>wrote:

> Hello Sumayya,
>
> Thank for you your interest in the Quranic Arabic Corpus, and
> specifically in the Ontology of Quranic Concepts. So far, we have
> received several requests for a downloadable copy of the Ontology, but
> I am afraid that I will have to decline, since this resource is not
> yet ready for full public release. Please allow me to explain why. The
> Quranic Arabic Corpus has three ongoing projects:
>
> - Morphology - This project involves annotating for part-of-speech,
> inflection features (e.g. person, gender, number) and other
> morphological features, including segmentation of prefixes and
> suffixes.
>
> - Syntax - This project involves assigning a syntactic function to
> each word in every sentence, and showing how words relate to each
> other through syntactic roles. Dependency grammar is used to visualize
> traditional Arabic syntax (i'raab).
>
> - Ontology - This more recent project is a step towards to the final
> aim of capturing some of the more straightforward semantics contained
> some the verses of the Quran. The ontology defines a list of base
> terms, and shows how these concepts relate using predicate logic.
>
> Although the morphology project is still ongoing, the data is of
> sufficiently high quality to merit a public release. As such, you can
> download the full morphological annotation of the Quran at
> http://corpus.quran.com/download. In contrast, the syntax project is
> ongoing. So far around 11,000 out of 77,430 words of the Quran are
> covered by dependency graphs, see:
> http://corpus.quran.com/treebank.jsp. The ontology project is still in
> its infancy. Although we have had some very good feedback about this
> last project, and many people do use it regularly, I still believe it
> has a long way to go.
>
> As I am sure you are aware, the important nature of the Quran as a
> central religious text sometimes requires a considered approach. It
> could be argued that as many as 1.5 billion people worldwide consider
> the Quran to be the literal word of God. In any case, I am very
> cautious and conservative with regards to any annotation of the Quran.
> With this in mind, the project is being run with the approach of
> supervised collaborative annotation. Improvements and suggestions are
> collected daily using an online message board forum:
> http://corpus.quran.com/messageboard.jsp. This high level of
> visibility ensures that potential mistakes are found and corrected, or
> at least are discussed in an open manner, where different opinions and
> views can be heard.
>
> I am reluctant to release any data from the Quranic Arabic Corpus for
> download until it is ready, to a high level of accuracy. Eventually,
> the plan is to release all data free of charge under the GNU public
> license. So far, the morphological annotation is in my opinion of
> sufficiently high quality to merit a public release. Eventually, the
> other datasets (the Syntactic Treebank and the Ontology) will be
> expanded to cover the entire Quran, and also release free of charge -
> hopefully later this year.
>
> However, having said that, I find your project very interesting. You
> say that you have been working on this project for the last 5 months?
> I would be interested to learn more about your project. What kind of
> data have you produced so far? How far have you got with verse
> similarly? It would be great to know more about how you are getting on
> with this. I would like to help - I would suggest you might look into
> the following:
>
> 1) Noorhan Abbas' Quran Topic Research -
> http://quranytopics.appspot.com (also at the University of Leeds). If
> you contact her directly, you may be able to ask her for some of her
> data.
> 2) Abdul-Baquee M. Sharaf is also working on verse similarly. See:
> http://www.comp.leeds.ac.uk/scsams
> 3) I have found that there are many good existing topic indexes of the
> Quran, although not often in machine readable form. For example, see:
> http://www.harunyahya.com/indexa.php
>
> I would be interested to know more about the knowledge models you have
> developed - how do these work?
>
> Kind Regards,
>
> - Kais Dukes
>
> Language Research Group
> School of Computing
> University of Leeds
>
> http://corpus.quran.com - The Quranic Arabic Corpus
> comp-quran@comp.leeds.ac.uk - Computational Quranic Arabic discussion list
>
>
> > On Sun, Mar 7, 2010 at 11:07 AM, Kais Dukes <s...@leeds.ac.uk> wrote:
> >
> > -------------------------------------------
> > From: sumayya baqai[SMTP:sumar...@gmail.com <smtp%3asumar...@gmail.com>]
> > Sent: Sunday, March 07, 2010 11:06:59 AM
> > To: Kais Dukes
> >
> > Respected Sir/ Madam,
> >
> > I hope this e-mail finds you in the best of your health.
> >
> > What brings me to email you is your marvellous work regarding Quran
> Arabic Corpus, So first of all let me congratulate you on such an excellent
> effort.
> >
> > I am Sumayya Rasheed Baqai, a student of final year of BS(CS) at FAST
> National University of Computer and Emerging Sciences, Islamabad, Pakistan.
> This year we had to undertake our Final Year Projects, and so the project I
> took is called Rabt-ul-Quran- with the aim of facilitating religious
> learning by modelling the thematically coherent structure and knowledge of
> the Holy Qur’an. The vision behind the execution of Rabt-ul-Quran is to
> represent the internal relationships between the verses of The Holy Quran
> and the inter-Contextual Relationships between the Quran and the Ahadith.
> For this purpose we are using Semantic Web Technology for efficient and
> effective knowledge retrieval, primarily, with the use of ontological models
> for the Holy Quran and Ahadith.
> >
> > Me and my group members have been working on this project for the past 5
> months under the supervision of Ms Amna Basarat Haider, Lecturer FAST-NU.
> >
> > So far, we have developed the knowldege models for the Holy Quran and
> Ahadith. The problem we are facing now is of lack of authentic data,
> categorised on the basis of theme, that relates the verses with the
> respective Ahadith. Recently, I saw your website that identifies themes
> using a well defined concept hierarchy.
> >
> > To fulfill our vision of Rabt-ul-Quran, we desperately need a hierarchy
> of concepts, one that helps us in identifying the basic themes from every
> verse. If possible, and if it does not violate any of your work policies,
> can you please share with us the Quran ontology that you have developed so
> far. I went through the index and was amazed to see its relevance from our
> project.
> >
> > Given the cause at hand, I firmly believe that any form of knowledge
> sharing would be a great service to the world, a benefit to the pool of
> Learning Sources for the religious Literature, for the generations to come.
> > I hope you will agree and I will look forward to receiving your support
> in fullfilling the vision.
> >
> > Thank you very much,
> >
> > Best Regards,
> > Sumayya Rasheeb Baqai,
> > FAST-NU
>
>


-- 
Best regards,
Abdul Baqi

Reply via email to