Testing on variables

42 views
Skip to first unread message

Daniel Stein

unread,
Mar 25, 2014, 5:26:24 AM3/25/14
to unitex-...@googlegroups.com
Dear all,

I have a question regarding testing on variables. Is it possible to use variables in order to test, if a subgraph was called or not?

I tried several alternatives but all I can achive is a concordance with each entry twice, one stating "TRUE" and the other one "FALSE". After many iterations my graph actually looks something like the one on the screenshot.

Thanks in advance

Daniel

Daniel Stein

unread,
Mar 25, 2014, 5:27:12 AM3/25/14
to unitex-...@googlegroups.com
ScreenShot006.jpg

Denis Maurel

unread,
Mar 25, 2014, 10:55:03 AM3/25/14
to Daniel Stein, unitex-...@googlegroups.com


Hi Daniel,

If you put other boxes after tests, you can do something if the subgraph is called or not.

Best regards,

Denis Maurel


____________________________________
Professor Denis Maurel
Université François Rabelais Tours
LI (Computer Science Research Laboratory)
EPU-DI
64 avenue Jean-Portalis
37200 Tours
France
Phone: 33-2.47.36.14.35
Fax: 33-2.47.36.14.22
mailto:denis....@univ-tours.fr

http://www.univ-tours.fr/maurel

http://www.li.univ-tours.fr
http://tln.li.univ-tours.fr/





Am Dienstag, 25. März 2014 10:26:24 UTC+1 schrieb Daniel Stein:


--
You received this message because you are subscribed to the Google Groups "Unitex-GramLab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unitex-gramla...@googlegroups.com.
To post to this group, send email to unitex-...@googlegroups.com.
Visit this group at http://groups.google.com/group/unitex-gramlab.
To view this discussion on the web visit https://groups.google.com/d/msgid/unitex-gramlab/74263826-170b-40e9-ac5b-0949df663cea%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Daniel Stein

unread,
Mar 25, 2014, 12:42:06 PM3/25/14
to unitex-...@googlegroups.com, Daniel Stein, denis....@univ-tours.fr
Hi and thanks again Denis,

although I am pretty sure I tested this before - you are absolutly right :-)

Kind regards
Daniel

Daniel Stein

unread,
Mar 26, 2014, 3:08:57 AM3/26/14
to unitex-...@googlegroups.com, Daniel Stein, denis....@univ-tours.fr
I tested it indeed, but now I know why it didn't work out: It was due to the use of a left context. I will tell you first what I want to achieve:

I want to test, if there is a determiner before a nominal construction or not. In order to do so I used

a) testing of a input-variable (before the nominal construction). This works but the recognized determiner now is a part of the output which I do not want. Results in 4659 matches.

b) placing a * left context right after the input-variable and before the test. This results in no longer showing the desired output. I do not understand this behavior as Unitex (as far as I understand it) should keep the value of a variable which is used inside of the left context.

c) placing a * left context right after the test. The results are the same as in b)

d) making the output of the test a output variable. I am not sure if this is possible due to the policy that processes outputs first but this special output relates with input that has to been processed? Well, as a matter of fact, the results are again the same as in b)

I am pretty sure that this should work but I am running out of ideas. Thanks in advance

Daniel
Reply all
Reply to author
Forward
0 new messages