#21853This paper describes an implemented program that takes atagged text corpus and generates a partial list of the subcategorization frames in which each verb occurs.
other,19-1-H91-1067,ak
and generates a partial list of the
<term>
subcategorization frames
</term>
in which each
<term>
verb
</term>
occurs
#21863This paper describes an implemented program that takes a tagged text corpus and generates a partial list of thesubcategorization frames in which each verb occurs.
other,24-1-H91-1067,ak
subcategorization frames
</term>
in which each
<term>
verb
</term>
occurs . The completeness of the
#21868This paper describes an implemented program that takes a tagged text corpus and generates a partial list of the subcategorization frames in which eachverb occurs.
other,11-2-H91-1067,ak
increases monotonically with the total
<term>
occurrences
</term>
of each
<term>
verb
</term>
in the
<term>
#21882The completeness of the output list increases monotonically with the totaloccurrences of each verb in the training corpus.
other,14-2-H91-1067,ak
total
<term>
occurrences
</term>
of each
<term>
verb
</term>
in the
<term>
training corpus
</term>
#21885The completeness of the output list increases monotonically with the total occurrences of eachverb in the training corpus.
lr,17-2-H91-1067,ak
</term>
of each
<term>
verb
</term>
in the
<term>
training corpus
</term>
.
<term>
False positive rates
</term>
#21888The completeness of the output list increases monotonically with the total occurrences of each verb in thetraining corpus.
measure(ment),0-3-H91-1067,ak
in the
<term>
training corpus
</term>
.
<term>
False positive rates
</term>
are one to three percent . Five
<term>
#21891The completeness of the output list increases monotonically with the total occurrences of each verb in the training corpus.False positive rates are one to three percent.
other,1-4-H91-1067,ak
</term>
are one to three percent . Five
<term>
subcategorization frames
</term>
are currently detected and we foresee
#21901Fivesubcategorization frames are currently detected and we foresee no impediment to detecting many more.
lr,8-5-H91-1067,ak
Ultimately , we expect to provide a large
<term>
subcategorization dictionary
</term>
to the
<term>
NLP community
</term>
and
#21924Ultimately, we expect to provide a largesubcategorization dictionary to the NLP community and to train dictionaries for specific corpora.
other,12-5-H91-1067,ak
subcategorization dictionary
</term>
to the
<term>
NLP community
</term>
and to train
<term>
dictionaries
</term>
#21928Ultimately, we expect to provide a large subcategorization dictionary to theNLP community and to train dictionaries for specific corpora.
lr,17-5-H91-1067,ak
<term>
NLP community
</term>
and to train
<term>
dictionaries
</term>
for specific
<term>
corpora
</term>
.
#21933Ultimately, we expect to provide a large subcategorization dictionary to the NLP community and to traindictionaries for specific corpora.