Searching for syntactic contexts

[Leonid Spektor]

Lara,

    As best as I understand your goal, I think you can accomplish it with
the following two commands:

    combo +t%mor +s"*|think*^(det|*+n|*)" +d3 +f *.cha
    freqpos *.cmb.cex

I have used a small sample file and I search for different combination of
two words, but the principle is the same. Here is an example output I got:

combo +t%mor +s"*|think*^(pro:dem|*+prep|*)" +d3 test.cha
****************************************
>From file <test.cha>
@Comment:    -----------------------------------
@Comment:    *** File "test.cha": line 292;
*PAR:              
               v|think pro:dem|that
@Comment:    -----------------------------------
@Comment:    *** File "test.cha": line 358;
*PAR:              
                           v|think&PAST prep|of
@Comment:    -----------------------------------
@Comment:    *** File "test.cha": line 416;
*PAR:              
               v|think pro:dem|that

freqpos test.cmb.cex
Fri Mar 19 15:16:59 2010
freqpos (18-Mar-2010) is conducting analyses on:
  ALL speaker tiers
****************************************
>From file <test.cmb.cex>
  1  prep|of             initial =  0, final =  1, other =  0, one word =  0
  2  prodem|that     initial =  0, final =  2, other =  0, one word =  0
  2  v|think             initial =  2, final =  0, other =  0, one word =  0
  1  v|think&past     initial =  1, final =  0, other =  0, one word =  0

Number of words in an initial position =  3
Number of words in an other position   =  0
Number of words in a final position    =  3
Number of one word utterences          =  0

freqpos will give you more information then you need, but just look for the
words in "initial" position. Alternatively you can run freq on the output
from combo.

Leonid.

On 18-03-10 17:30, "Lara" <lara.klainerman at gmail.com> wrote:

> Hi,
> 
> I'm trying to analyze the syntactic patterning of the mental state
> verbs think, know, remember, and wonder in child-directed speech. In
> other words, I want to see how often they occur in particular
> syntactic frames (ultimately, to see whether these verbs are
> differentiated by their syntactic patterns).
> 
> For example, I want to find out how often each verb is followed by
> "about" (e.g. "He's thinking about the story"), how often each verb
> takes an NP complement (e.g. "I know the story"), and how often each
> verb takes a VP-ing complement (e.g. I remember playing...").
> 
> Occurrences of a verb plus another word (e.g. think + about) shouldn't
> be too difficult to search for -- I can just use COMBO. However, COMBO
> doesn't seem to allow +r for searching stems?
> 
> More importantly, though - is it possible to search for abstract
> elements like NP and VP? I know the %mor tier only codes parts of
> speech, but I was hoping I could make a file with a list of all
> possible NP breakdowns (det|* n|*, n|*, etc.) and do a freq search for
> "think" followed by any element of that list? Something like:
> 
> freq +s"think"^@nounphrases.cha
> 
> I've tried to do this and have failed; I don't know whether it's
> because it can't be done (e.g. because you can't use wildcards or two
> words within a list in a separate file) or because I'm doing something
> wrong.
> 
> Any help would be greatly appreciated! Thank you so, so much in
> advance.
> 
> Lara


-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To post to this group, send email to chibolts at googlegroups.com.
To unsubscribe from this group, send email to chibolts+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/chibolts?hl=en.