Chart dependencies [Was: [developers] active/passive parser bug]

Berthold Crysmann crysmann at dfki.de
Wed May 4 10:43:50 CEST 2005


Hi,

while debugging my lkb-spelling code for German separable particle 
verbs,  I just wondered what the policy will be for chart dependencies 
with the new approach to lexical rules. In PET (at least in Bernd's 
version) it is possible to configure whether dependencies should be 
checked before or after processing of lexical rules.

There is a concrete issue for German: in my current version of the 
grammar I derive the nonseparated lexical entry from the separated one 
by means of a morphological rule. Main branch LKB appears to check for  
chart dependencies after application of spelling rules, which gives the 
desired results. As Bernd assures me, this is also the behavior in PET 
(though I will check this later today).
Application before any rules, however, will be pretty useless for German.


Berthold


Stephan Oepen wrote:

>hi ben,
>
>  
>
>>Items #239 and #241 of the CSLI test suite produce many more parses when 
>>processed in passive rather than active mode. This would appear to be a 
>>bug.
>>    
>>
>
>it is in principle possible for the two parsers to differ in the number
>of results found, albeit not desirable.  the active parser includes the
>`chart dependency' filter originally proposed by dr. mueller (for all i
>know, that is).  checking `chart dependencies' prior to parsing aims to
>suppress search doomed to fail, but of course there is room here for a
>grammar to go astray and accidentally suppress intended analyses.  one
>way of testing this hypothesis would be forcing 
>
>  (defparameter *chart-dependencies*
>    '((SYNSEM LKEYS --+COMPKEY) (SYNSEM LOCAL CAT HEAD KEYS KEY)
>      (SYNSEM LKEYS --+OCOMPKEY) (SYNSEM LOCAL CAT HEAD KEYS KEY)
>      (SYNSEM LKEYS --+SUBJIND) (SYNSEM --SIND)))
>
>to nil after loading the ERG.  --- i just tried PET and get:
>
>  (1) `Abrams sees Browne as being a competent manager .' [0] --- 2 [...]
>
>so at least ulrich and i are on the same side here :-).
>
>                                                         best  -  oe
>
>+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>+++ Universitetet i Oslo (ILN); Boks 1102 Blindern; 0317 Oslo; (+47) 2285 7989
>+++     CSLI Stanford; Ventura Hall; Stanford, CA 94305; (+1 650) 723 0515
>+++       --- oe at csli.stanford.edu; oe at hf.uio.no; stephan at oepen.net ---
>+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>  
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.delph-in.net/archives/developers/attachments/20050504/1416ee57/attachment.html>


More information about the developers mailing list