These are chat archives for cltk/cltk

7th
Mar 2016
James Tauber
@jtauber
Mar 07 2016 12:15
@lukehollis I'm working a reading environment related to the data @jonathanrobie linked to (and an API I wouldn't mind aligning more with your work)
@jonathanrobie great to see you here!
Luke Hollis
@lukehollis
Mar 07 2016 14:06
@jtauber that's great! would love to learn more about it
Nathan D. Smith
@nathans
Mar 07 2016 16:51
Hi folks. I work in the ancient Greek nlp space. Currently studying cltk and will see if there is anything I can contribute.
Kyle P. Johnson
@kylepjohnson
Mar 07 2016 16:53
@nathans Always happy to have news eyes and thoughts on the project. Let us know how you think it can be improved.
Nathan D. Smith
@nathans
Mar 07 2016 17:00
@kylepjohnson with respect to elasticsearch and nlp, have you seen the xtas project? http://xtas.net/index.html
Kyle P. Johnson
@kylepjohnson
Mar 07 2016 17:04
That's interesting. Do you know what kinds of problems xtas is trying to solve? For us, I think a good ES indexer isn't out of he question. I wrote one once for the TLG but decided it wasn't worth the overhead. Then again, now that Luke is working on a CLTK website, I'm thinking we will need something that will index all corpora hosted by CLTK.
Nathan D. Smith
@nathans
Mar 07 2016 17:07
I like elasticsearch (not necessary coupled with xtas) due to the turnkey advanced search capabality (e.g. for advanced syntax queries, etc.) with a nice API.
I happen to use the product in my professional life (totally different application) and have a fondness for it
Kyle P. Johnson
@kylepjohnson
Mar 07 2016 17:11
So cool. @nathans would it be alright to ping you when we need to set one up for the website? I know the basics, but I know how important the many parameters are.
evezhier
@evezhier
Mar 07 2016 17:25
Hi all, I am a student in NLP and I want to extend CLTK to a new language. Should I continue here for details or by email?
Kyle P. Johnson
@kylepjohnson
Mar 07 2016 17:27
@evezhier please email me (address on my github) with which languages, what you will add, and what data sets you'll use. Thanks!