What impact did Systemic Functional Language / Grammar have on natural language processing?

If anyone's interested in doing some work on NLP with Systemic Functional Grammar, I have a corpus conversion of the Penn Treebank into Introduction to Functional Grammar/Lexicogrammatical Cartography-style SFG annotation that I never really did anything with. The conversion's described in this paper:

@InProceedings{honnibal-curran:2007:ACL2007_DLP,
  author    = {Honnibal, Matthew  and  Curran, James R.},
  title     = {Creating a Systemic Functional Grammar Corpus from the Penn Treebank},
  booktitle = {ACL 2007 Workshop on Deep Linguistic Processing},
  month     = {June},
  year      = {2007},
  address   = {Prague, Czech Republic},
  publisher = {Association for Computational Linguistics},
  pages     = {89--96},
  url       = {http://www.aclweb.org/anthology/W/W07/W07-1212}
}

If you want to play with the data (and have a Penn Treebank 3 license) let me know. I did the conversion during my honours project with Christian Matthiessen in 2004, and have always felt a little bad that I didn't follow through on it much, past publishing about a slightly cleaned up version in 2007. It's always felt like the conversion could use just a little more work before release, as these things often do.

I think saying that systemic functional grammar adopts a "sociological" view of linguistics is a little narrow/imprecise. While there is a grain of truth in it (an interest in how language functions within a community of humans as users), the "systemic" part refers to choices in a formal system.

Systemic Functional Grammar has had the most impact in the area of Natural Language Generation. The idea of a system of choices to express a meaning is especially congenial to an NLG approach, more so than most formal linguistic systems, which tend to appear more suited to natural language analysis. Many of the best known NLG systems are inspired by or closely based on versions of SFL, including KOMET, KPML, PENMAN, and WAG, including work by people such as Mann, Bateman, and Matthiessen, but also crossing over to computational linguists more known in the formal linguistics community, such as Hovy, Kasper and Mellish.

I don't know about SFL myself, but I would start by searching in the ACL Anthology for "Systemic functional grammar". Here are the results.