TIGERSearch

TIGERSearch is a java-based (and therefore platform-independent) search engine for syntactically annotated corpora with an optional, highly developed graphical user interface. Currently, TIGERSearch is installed on AFS and on the Corpus Computer. We also have the installation files for windows, unix, and linux avaible on AFS (let the corpus TA know if you need the installation file for an other operating system). To learn more about the advantages and disadvantages of TIGERSearch (compared to tgrep) have a look at the presentation linked below under "Tip-1". In addition, this page deals with the following topics.

Tip-1: Before you start working with TIGERSearch, you may find it useful to have a look at the overview of search tools for syntactically annotated corpora available at Stanford (from the 02/06/2004 tutorial) available as pdf and as interactive html presentation.

Tip-2: For beginnners and people who are looking for a full-blown step-by-step introduction to the search syntax of TIGERSearch, there is the description of the TIGERSearch query language

Tip-3: If you already know the syntax or just want to get an overview, we recommend the TIGERSearch 2.1 quick reference to query language

Where to find TIGERSearch

TIGERSearch is installed on the Corpus Computer (simply log in open the start menu and you will find TIGERSearch listed among all the other programs) and on AFS. The installed program (TIGERSearch and the accompanying TIGERRegistry) are located at:

    /afs/ir/data/linguistic-data/TIGERSearch

In that directory you will also find a readme file with more information:

    /afs/ir/data/linguistic-data/TIGERSearch/00README.html

The corpus TA has signed the user agreement for TIGERSearch for ALL members of the department. So, as long as you follow the license conditions mentioned on their webpage you can also use our local installation file to install your own version locally. The Solaris, Linux, and Windows installation files are stored at:

    /afs/ir/data/linguistic-data/downloadables/

How to setup your account to use TIGERSearch on AFS

To make working with TIGERSearch more convenient you should change your PATH variable. Add the following line to the end of your .login file (note that this is slightly different from what was send out to the corpus list when the first TIGERSearch version was installed - both variant work):

    setenv PATH /afs/ir/data/linguistic-data/TIGERSearch/bin:$PATH

Your .login file is located in your home directory. You can get there and open the file by typing:

    cd ~
    emacs .login -nw

Within the emacs editor go to the bottom of the file (simply use the arrow keys), add the line mentioned above, and then type CONTROL-X (i.e. press the control key and at the same type 'x') and then CONTROL-S. This will save the change. You can leave the editor by typing CONTROL-X CONTROL-C.