Hello!

Please introduce yourself here, if you haven't already.
Post Reply
emilianofs
Posts: 4
Joined: July 21st, 2020, 9:57 am

Hello!

Post by emilianofs »

Hi everybody,

My name is Emiliano, I was born in Argentina and am currently living in Spain.

I came across Ancient Greek back in 2014 while studying philosophy. I took the first Ancient Greek Language module and then continued studying it by myself. First, I got caught up by Heraclito's fragments, then Plato and finally the NT & LXX.

Everyday I do my best to read some fragments of my Nestle-Aland NT. Since there is always a new unknown word or declension or conjugation, I am a heavy user of the Perseus Greek Morph Tool.

My field of work is IT and have been working with Product Management fucused on Cloud Computing for some time now. Currently I am doing a Master's Degree in Software Engineering. Next semester I will be working on my final project and I would like to do something related to Biblical Greek and AI.

For sure this Forum will be super helpful for me!

Thanks,
Emiliano Spinella
Barry Hofstetter
Posts: 2159
Joined: May 6th, 2011, 1:48 pm

Re: Hello!

Post by Barry Hofstetter »

Welcome to B-Greek!
N.E. Barry Hofstetter, M.A., Th.M.
Ph.D. Student U of FL
Instructor of Latin
Jack M. Barrack Hebrew Academy
καὶ σὺ τὸ σὸν ποιήσεις κἀγὼ τὸ ἐμόν. ἆρον τὸ σὸν καὶ ὕπαγε.
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Hello!

Post by Jonathan Robie »

Yes, welcome!

When you say "related to Greek and AI", are you thinking in terms of Natural Language Processing (NLP) or something else?
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
Daniel Semler
Posts: 315
Joined: February 18th, 2019, 7:45 pm

Re: Hello!

Post by Daniel Semler »

emilianofs wrote: July 21st, 2020, 10:22 am
My field of work is IT and have been working with Product Management fucused on Cloud Computing for some time now. Currently I am doing a Master's Degree in Software Engineering. Next semester I will be working on my final project and I would like to do something related to Biblical Greek and AI.

For sure this Forum will be super helpful for me!

Thanks,
One thing I thought might be fun to play with was using a classifier to auto-tag for various features. There are of course a number of autotaggers about. But I wondered if a neural net trained on a decent size text might do a better job. You don't have a really large training set available though alas. The NT has been worked over pretty well but not the LXX. But it might be interesting to see how well an NT trained tagger would work on the LXX.

Thx
D
emilianofs
Posts: 4
Joined: July 21st, 2020, 9:57 am

Re: Hello!

Post by emilianofs »

Jonathan Robie wrote: July 21st, 2020, 11:26 am Yes, welcome!

When you say "related to Greek and AI", are you thinking in terms of Natural Language Processing (NLP) or something else?
Hi Jonathan,

Yes, I've already done some utilities that resemble your Jupyter notebooks using XQuery. Basically, I've merged Nestle NT & LXX morphology into a unified data structure. So, for instance, I am now able to query for lemma frequencies in both corpora. However, I don't yet have a very clear idea what would be a good contribution to the community.

On the other hand, I am really interested in POS analysis. For instance, I would like to get all the predicates for a given noun. Imagine getting all the predicates for theós and comparing them between corpora. I would imagine that the Gospel of John would predicate agáph to theós much more in the NT than in the LXX. Nevertheless, I couldn't find a POS treebank for LXX and it would be completely impossible for me to do.

Then I also have another idea: mapping Revelations symbols to the LXX. For instance identifying all the symbols (candlesticks, stars, etc.) and getting their meaning from LXX.

Finally, I am also very interested in the hyphothesis that the main reason for early gentiles conversion was finding in Jesus Christ the fulfillment of the OT prophesies. Thus, another idea is to identify in the NT all the fullfilled prophesies and map them to the LXX.

As you can see, I am still elaborating and going through a lot of articles. However I cannot yet find the right material, not even a simple word classification analysis of the corpora.

Any ideas are welcome!

Best,
Emiliano Spinella
emilianofs
Posts: 4
Joined: July 21st, 2020, 9:57 am

Re: Hello!

Post by emilianofs »

Daniel Semler wrote: July 21st, 2020, 4:30 pm
emilianofs wrote: July 21st, 2020, 10:22 am
My field of work is IT and have been working with Product Management fucused on Cloud Computing for some time now. Currently I am doing a Master's Degree in Software Engineering. Next semester I will be working on my final project and I would like to do something related to Biblical Greek and AI.

For sure this Forum will be super helpful for me!

Thanks,
One thing I thought might be fun to play with was using a classifier to auto-tag for various features. There are of course a number of autotaggers about. But I wondered if a neural net trained on a decent size text might do a better job. You don't have a really large training set available though alas. The NT has been worked over pretty well but not the LXX. But it might be interesting to see how well an NT trained tagger would work on the LXX.

Thx
D

Yes, that sounds interesting. However, I've been reading LXX and the grammar/syntax is very heterogeneous. For instance, I found out by doing some analysis in the LXX that there are some books which do not even use punctuation! So there are full long chapters with only one sentence.

I am not an expert in ancient greek, not at all. But I imagine that the ~3 centuries time difference between LXX & NT are evident in the text.

Maybe one option can be training a CNN with all of the available treebanks and see how that works. Maybe it's an option for tagging the LXX. I'll give it a thought.

Thanks!!!
Emiliano Spinella
Jonathan Robie
Posts: 4158
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: Hello!

Post by Jonathan Robie »

emilianofs wrote: July 22nd, 2020, 5:47 am Yes, I've already done some utilities that resemble your Jupyter notebooks using XQuery. Basically, I've merged Nestle NT & LXX morphology into a unified data structure. So, for instance, I am now able to query for lemma frequencies in both corpora. However, I don't yet have a very clear idea what would be a good contribution to the community.

On the other hand, I am really interested in POS analysis. For instance, I would like to get all the predicates for a given noun. Imagine getting all the predicates for theós and comparing them between corpora. I would imagine that the Gospel of John would predicate agáph to theós much more in the NT than in the LXX. Nevertheless, I couldn't find a POS treebank for LXX and it would be completely impossible for me to do.
Right, you need something you don't have. GBI has a commercial treebank for LXX that has been licensed to Logos. It is not as good as their GNT treebank, but it is reasonably useful for the kind of thing you mention. But if you use it you are stuck in the position of being unable to share your data or build something that can be used by others. And to use it, you would have to get permission, which might be complicated.
emilianofs wrote: July 22nd, 2020, 5:47 am Then I also have another idea: mapping Revelations symbols to the LXX. For instance identifying all the symbols (candlesticks, stars, etc.) and getting their meaning from LXX.
For this, it might be useful to start with something like the Dictionary of Human-Made Things in the Bible, Dictionary of Plants and Trees in the bible, Dictionary of Animals in the Bible, etc. Or simply use Louw & Nida. If you were a translator on a translation project this would be easy ... there is a semantic domain crawler in Paratext. By this November a lot of this might be more widely available to people who are not translators. I might be able to negotiate permission for you. Of course, "getting their meaning from LXX" is complicated and involves a lot of human judgement. I don't think AI alone can do this for you. I'm not sure how helpful NLP or AI would be here at all.
emilianofs wrote: July 22nd, 2020, 5:47 am Finally, I am also very interested in the hypothesis that the main reason for early gentiles conversion was finding in Jesus Christ the fulfillment of the OT prophesies. Thus, another idea is to identify in the NT all the fullfilled prophesies and map them to the LXX.
Many people have done this by hand. Again, I'm not sure how AI or NLP would particularly help, or how you would tell the engine where to look for the fulfilment of these prophecies. If you do this automagically, I also suspect you might frequently miss the point unless you take figures and types into account properly. I would read the last three books by Richard Hays before starting anything like this if I were you.
emilianofs wrote: July 22nd, 2020, 5:47 am As you can see, I am still elaborating and going through a lot of articles. However I cannot yet find the right material, not even a simple word classification analysis of the corpora.
If you PM me, we can set up a conversation. It might be easier to steer you in the right direction by talking first.
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/
Post Reply

Return to “Introductions”