Ability Text Grammar Induction

[YeGoblynQueenne]

When I was working on Incantus, I wound up spending some time making the oracle parser create ability code automatically, by parsing the text.

This code, as far as I can tell, never made it into any of the source archives or repositories, because it was part of the web editor, the source for which was not released (and currently doesn't seem to be working on my site; probably due to Django updates).

So I thought I'd put it on pastebin here: http://pastebin.com/FMB7Xaqy

While this is rather out of date (last updated at the end of August in 2012, and Magic templating has changed quite a bit since then), it did manage to parse a significant number of cards without need of human input to correct it. It's possible somebody might find it useful for something, so I don't care what you do with it as long as you credit Incantus and myself if you do anything based on it.

Ohi MageKing

Yep, I remember your parser also. That (and any other parser or any other relevant M:tG program I can find) is getting a reference in my dissertation, as prior work (that's a good thing eh?). Since it's an academic project I'd be absolutely required to cite it as your work if I were to use it anyway, so no worries about that. I ain't using your work without citing- that would be plagiarism and my ass would be grass :/

The thing is, I kind of moved away from the supervised approach in the last few weeks and now I have a program that can learn a grammar from examples, without any prior knowledge - it doesn't need an existing grammar anymore. Today in fact I just had a bit of a breakthrough as I realised what I need to do to make the grammar a fully hierarchical grammar (rather than just a set of fragments of a grammar which is what I had until now).

I'll keep you guys updated once I have something that others beside me can use.

[Arcanist]

I'll keep you guys updated once I have something that others beside me can use.

While replying to a different thread I realized today that I kind of forgot about this thread. Which is too bad because it was rather interesting. If you are still around, did you ever write your dissertation? What came out of your experiments with grammar induction? I would be very interested in reading about it.

[YeGoblynQueenne]

I'll keep you guys updated once I have something that others beside me can use.

While replying to a different thread I realized today that I kind of forgot about this thread. Which is too bad because it was rather interesting. If you are still around, did you ever write your dissertation? What came out of your experiments with grammar induction? I would be very interested in reading about it.

Hi. Apologies for the late reply, I should have checked this thread a bit more often.

I did finish my project, yes, and got my Masters. In the end, I didn't focus too much on creating an AT parser. This was an MSc thesis and so I had limited time to work on it. I did create a novel grammar induction algorithm though and generated several types of AT grammar using it.

The latest code for the project can be found on my public git repos, here:

http://goblinopera.com:1234/?p=at_gramm ... ;a=summary

My dissertation report is also hosted on my server, here:

http://goblinopera.com/THELEMA/THELEMA_report.pdf

I understand this may sound a bit dodgy (the personal git server I mean. It was done as a demonstration for an employer). There's also a github repo, but both the code and particularly the documentation on it are a bit outdated:

https://github.com/stassa/THELEMA

Note that you probablly won't be able to get the project up and running on your own- my documentation is not very good. I'm happy to help if required- just throw me a line on eIA!patsantzis17CTHULHUimperialIA!acIA!uk, replacing "CTHULU" with "@" and "IA!" with ".".

I'm now doing an AI PhD and I'm still interested in M:tG as a research subject, though not only with respect to the AT language. I'll have to do it all in my own time though, because I probably won't get any funding for it

[Arcanist]

Hi. Apologies for the late reply, I should have checked this thread a bit more often.

Thanks for the reply. I was checking the forum regularly a year ago. Then I got sick, and had to make up for lost time at work. Since then I have benn very busy and I have not had much time to connect. I was randomly looking at this thread today when I saw your message.

I have started looking at your thesis (I have only read the introduction, conclusion, and bibliography so far ). It looks interesting. Not sure I will be able to make much use of the code though. My Prolog is very rusty, and I never really used that language for any advanced programs to begin with.

I will probably drop you a line at your email address.