[sword-devel] Strong's Numbers Python Script for Foreign Language Bibles

[jonathon]

On 28/06/2011 16:52, Kamal Abou Mikhael wrote:

> 1. You mentioned the script being tied to Bibletech talk... was that your
 talk?

No.

>are there any remaining artifacts from that talk?  mp3... slides...
notes...

They should be on the BibleTech website.

> 2. How does the script work?  Does it use advanced computational linguistics
 of some type or heuristics of some kind?

Basically brute force.
It has four required files attached to it.
* Hebrew Strong Numbers, in Hebrew Word Order, for the entire Bible. (NT
& MT);
* Greek Strong Numbers, in Greek Word Order, for the entire Bible. (LXX
& NT);
* English word frequency/Strong Number/scripture citation combinations;
* English text;
The fifth file is the Bible translation in the target language.

The script reads the target language.
* Create a concordance of the Bible in the target language;
* Create a word frequency chart in the target language;
* Start with the first word of the first verse:
# Locate that word in other verses where the same Strong Number appears;
# If matches are found, then the Strong Number is attached to the word;
# If a mismatch is found, that gets notated with (?);
# If no matches are found, then the word is skipped;
# At the end of the verse, left over Strong Numbers are put in
parenthesis at the end of the verse;
# At the end of the session, it rereads the verses, looking for unmarked
words, and Strong Numbers at the end of the verse;
# It then attempts to rematch words, dropping syllables to locate the match;

Messy, and only useful for a first cut rough draft.

I went through the most probable set of DVDs and CDs that I would have
archived it in, and didn't find the script.  :(

jonathon
-- 
All emails sent to this with email address with a precedence other than
bulk, or list, are forwarded to Dave Null, unread.