Usage of default parsers

If we want to use the default parsers included in the package, the command to be use is the following:

dp.sh  -a treetagger en mytext.txt > mytext.dep

This command generates a simple dependency-based representation (option -a), making use of the english (en) treetagger. The expected result is:

SENT::<I_PRO_0_<number:0|lemma:I|possessor:0|case:0|genre:0|person:0|politeness:0|type:P|token:I|>
have_VERB_1_<number:0|mode:0|lemma:have|genre:0|tense:0|person:0|type:A|token:have|>
a_DT_2_<number:0|lemma:a|possessor:0|genre:0|person:0|type:0|token:a|>
dream_NOUN_3_<number:S|lemma:dream|genre:0|person:3|type:C|token:dream|>._SENT>
(SubjL;have_VERB_1;I_PRO_0)
(SpecL;dream_NOUN_3;a_DT_2)
(DobjR;have_VERB_1;dream_NOUN_3)


The output consists of 4 lines. The first one is the input of the parsing strategy. It starts by 'SENT::' and contains 5 columns separated by a space. Each column is provided with the morpho-syntactic information assigned to each token of the input text. The remaining 3 lines show the dependency-based analysis. For instance, the second line represents the Subject dependency between pronoun ``I'' and verb ``have''.

If we would like to parse a spanish text, we should use the following command:

dp.sh  -a treetagger es mytext-es.txt > mytext-es.dep

where 'mytext-es' stands for the name of a file containing a text in Spanish. The remaining languages are called with ``fr'' (french), ``pt'' (portuguese), and ``gl'' (galician).

Instead of treetagger, we can use freeling:

dp.sh  -a freeling en mytext.txt > mytext.dep

Freeling is not provided with the DepPattern package. It must be previously installed. The configuration files should be in '/usr/local/share/FreeLing/config/', which is the by default directory in the standard installation.

Pablo Gamallo 2009-10-02