I have used computational linguistics approaches to learn and classify the psyche of the author in Old Testament and New Testament and shown detailed analysis and results in text visualization form in my book Text Psyche Mining which is available with Amazon, EBay and Google. Using simplistic association analysis of keywords is not much effective in relating concepts discussed in various verses unless there is some ontological classification of topic. However, its good for a beginning.
Ahsan Nabi Khan
On Wed, Jul 10, 2013 at 10:45 AM, Tariq Mahmood <tariq.mahmood@nu.edu.pk> wrote:
Assalamoalaikum,A relatively novel research field is related to the application of data mining techniques on Holy scriptures, e.g., Holy Qur'an, Holy Bible, Holy Torah etc. From Qur'anic perspective, the field is coming to be known as Qur'an Mining. This paper published in 2010 provided the first direction on this field. The aim is to extract interesting patterns which can provide useful insights to facilitate and support the interpretational works on the Qur'an, along with providing interesting knowledge at high level of abstraction for the general population. This is another ongoing work by a PhD student in Leeds.Recently I have done some work in Qur'an Mining with one of my FYP groups, using the Rapid Miner tool. The project website is https://sites.google.com/site/miningthequran/. We have applied cluster analysis (through CLOPE algorithm) on Qur'anic verses of all 30 chapters. Through this, we extracted the "significant" topics mentioned in each chapter (CLOPE calculates this significance based on the frequency of occurrence of similar values - more similar values across rows imply greater significance). In all, we extracted 226 topics, e.g., Allah, Day of Judgement, Aad, Thamud, Saba Valley etc. The topics for each chapter (juzz) are shown in https://sites.google.com/site/miningthequran/text-mining-of-qur-an/graphsThese topics act as types of "bookmarks" or keywords for each chapter, e.g., the the topic "Divorce" is significant in Chapter 28 in which all matters pertaining to divorce are explained in Surah Tallaq.As a next step, we performed association rule mining to extract probabilistic associations between the extracted topics. For instance the rule Aad 13 --> Thamud 13 (Confidence: 1) implies that in 13 chapters, whenever Aad has been clustered then Thamud has also been clustered. This leads us to question: "Why Aad and Thamud are significant (mentioned) collectively in this way?" - Scholars have in fact responded: "Both were disbelieving nations, with Thamud being the relative of Aad, and coming in time immediately after Aad; both received severe exemplary punishments from God for not following their Prophets". Other rules can be viewed at https://sites.google.com/site/miningthequran/text-mining-of-qur-an/association-rule-mining.
We are still working to improve our cluster analysis and the rule mining. I would like to invite anyone who has any comments, suggestions or idea regarding this work to send me a private email.Regards--Dr. Tariq Mahmood
Assistant Professor - Coordinator Graduate Students CommitteeFAST- NU, Karachi Campus, PakistanTel: 111 128 128 - Ext: 233
__._,_.___
| Reply via web post | Reply to sender | Reply to group | Start a New Topic | Messages in this topic (2) |
.
__,_._,___
No comments:
Post a Comment