Comments on: The State of the Koine Parser as of October 2008 http://www.julian.textcrit.com/?p=17 An assortment of NT textual criticism comments and, frequently, how it relates to computers... Wed, 08 Sep 2010 02:08:19 +0000 http://wordpress.org/?v=2.3 By: Stephen C. Carlson http://www.julian.textcrit.com/?p=17#comment-11364 Stephen C. Carlson Fri, 14 Nov 2008 15:58:32 +0000 http://www.julian.textcrit.com/?p=17#comment-11364 Great to hear that you'll be at SBL in Boston. Since we tend to inhabit the same sessions, I'm sure we'll get a chance to catch up. Great to hear that you’ll be at SBL in Boston. Since we tend to inhabit the same sessions, I’m sure we’ll get a chance to catch up.

]]> By: Mike http://www.julian.textcrit.com/?p=17#comment-11248 Mike Wed, 29 Oct 2008 04:52:56 +0000 http://www.julian.textcrit.com/?p=17#comment-11248 The fact that LFG requires a lot of morpho-syntactic & semantic information might be a burden. From what I've read about computational linguistics and LFG, the majority has use a probabilistic approach, though I am not a computational linguistic, so I can't say more. You could search through the presentation titles and abstracts at their conferenc website. A large number of papers are freely available there. http://www.essex.ac.uk/linguistics/lfg/FAQ/conferences.html The fact that LFG requires a lot of morpho-syntactic & semantic information might be a burden. From what I’ve read about computational linguistics and LFG, the majority has use a probabilistic approach, though I am not a computational linguistic, so I can’t say more. You could search through the presentation titles and abstracts at their conferenc website. A large number of papers are freely available there.

http://www.essex.ac.uk/linguistics/lfg/FAQ/conferences.html

]]> By: theswain http://www.julian.textcrit.com/?p=17#comment-11237 theswain Tue, 28 Oct 2008 04:17:49 +0000 http://www.julian.textcrit.com/?p=17#comment-11237 Just a note to say its good to have you back blogging. Not nearly enough blogs on text crit about. Just a note to say its good to have you back blogging. Not nearly enough blogs on text crit about.

]]> By: Julian http://www.julian.textcrit.com/?p=17#comment-11216 Julian Sat, 25 Oct 2008 18:57:18 +0000 http://www.julian.textcrit.com/?p=17#comment-11216 I remember reading a very long paper on LFG by Bresnan (?) and thinking that its notional scheme was very useful for Greek but that it didn't offer any significant advantage over a traditional tree structure. The main problem I have encountered is determining the local phrase head as I iterate over the sentence. Maybe you could recommend some reading on that topic. I am familiar with how LFG ensures grammatical matches through its nested feature structures but how does it solve ambiguity? Is there a probabilistic approach? And if so, how is this established without a corpus? I also agree that Greek is not all that free, there are certainly many constructs that are formulaic. I relied heavily on Wallace's Biblical Greek Beyond the Basics for comprehensive overviews of constructs. If I had a corpus, I could have compiled a better overview. Julian I remember reading a very long paper on LFG by Bresnan (?) and thinking that its notional scheme was very useful for Greek but that it didn’t offer any significant advantage over a traditional tree structure. The main problem I have encountered is determining the local phrase head as I iterate over the sentence. Maybe you could recommend some reading on that topic. I am familiar with how LFG ensures grammatical matches through its nested feature structures but how does it solve ambiguity? Is there a probabilistic approach? And if so, how is this established without a corpus?

I also agree that Greek is not all that free, there are certainly many constructs that are formulaic. I relied heavily on Wallace’s Biblical Greek Beyond the Basics for comprehensive overviews of constructs. If I had a corpus, I could have compiled a better overview.

Julian

]]> By: Mike http://www.julian.textcrit.com/?p=17#comment-11214 Mike Sat, 25 Oct 2008 17:15:44 +0000 http://www.julian.textcrit.com/?p=17#comment-11214 I don't think you looked closely enough at LFG. Much of the frameworks foundational work is based on non-configurational languages - such as the Austronesian Walipiri. It is actually highly suited for free word order languages (a good amount of LFG work has been done on Modern Greek), though I would argue that Greek is not as free as some would think. I'm presently writing an LFG grammar for Koine morphology & phrase structure using SIL software. I don’t think you looked closely enough at LFG. Much of the frameworks foundational work is based on non-configurational languages - such as the Austronesian Walipiri. It is actually highly suited for free word order languages (a good amount of LFG work has been done on Modern Greek), though I would argue that Greek is not as free as some would think. I’m presently writing an LFG grammar for Koine morphology & phrase structure using SIL software.

]]>