This tutorial will have three components: Unix, TGrep2 and the TGrep2 Database Tools. It will take place in the HLPlab (Meliora 123) from 10:30am - 12pm on the indicated Tuesdays and Thursdays. Depending on your background, you may not want to attend all of the sessions. To use TDTlite you'll need to know how to use TGrep2 which in turn requires that you be able to navigate in a shell. We're devoting one session to learning how to use a command line (for Unix newbies), two sessions to TGrep2, and two sessions to TDTlite.

Unix primer: October 22 (Judith)

To use TGrep2 and TDTlite you need to be able to navigate on a command line. If you don't know how to do this, you should come to this primer. Here's a cheatsheet containing some of the commands we'll go through: attachment:cheatsheet.txt

Introduction to TGrep2: October 27/29 (Florian)

Using TDTlite requires knowledge of TGrep2, which will be taught in two sessions. TGrep2 is a query language that allows you to search syntactically annotated corpora via regular expression syntax. The [http://tedlab.mit.edu/~dr/Tgrep2/tgrep2.pdf TGrep2 User Manual] is a handy document to have with you on this journey. Here are the slides from the first session: attachment:tgrep2_slides.pdf

Introduction to the TGrep2 Database Tools (TDT): November 3/5 (Judith)

In these two sessions we'll get into how to use TDTlite, a collection of scripts that not only lets you run TGrep2 queries in batch mode, but that lets you combine the output of those queries into a database in a variety of user-defined ways (i.e. you can add extracted strings themselves, frequency information, part-of-speech information, phonological information, length information, etc). TDTlite output is a tab-separated database that can be imported into your favorite stats program for further analysis. Here is the preliminary [wiki:HlpLab/TDT TDT User Manual] that will be transformed into a tutorial with example shortly.

MoinMoin Appliance - Powered by TurnKey Linux