For execute the system you need:

• Download IRASubcat zip IRASubcat.zip or the tar version IRASubcat.tar

• Uncompress the file. It is going to create a folder named IRASubcat.
  

• Put your corpus(for example corpus.xml) into the folder IRASubcat. It need to be XML file in UTF-8 format.

• Identify morphosintactic characterist and value for verbs into the corpus (for example sint=V).

• Identify the element XML father of element which have the characteristic to study (for example sentence).

• Identify characteristic or lexical items that do you want to have the values in the output dictionary for keys (example lexical or lema).

• If you have a existing dictionary, verbal list, or configuration file put them into the folder IRASubcat. If you don't put a config.cfg the system is going to run with the default configuration. If you are going to consider a list of verb, you need to put a file with them. They need to be UTF-8 format.

• and execute in comands line with the promp inside of IRASubcat folder:

>python IRASubcat.py corpus.xml sint=V sentence lexical (with corpus.xml example 1,2)

>python IRASubcat.py corpus.xml sint=V sentence lema (with corpus.xml example 2)

>python IRASubcat.py corpus.xml sint=V sentence lexical config.cfg (with corpus.xml example 1,2)

>python IRASubcat.py corpus.xml sint=V sentence lema config.cfg (with corpus.xml example 2)

             Example corpus.xml (1)

<corpus>

<sentence ID='1'>

<word>They</word>

<word sint='V'>are</word>

<word sint='V'>playing</word>

</sentence>

  </corpus>

              Example corpus.xml (2)

  <corpus>

<sentence ID='1'>

<word lema='they' sint='pron'>They</word>

<word lema='be' sint='V'>are</word>

<word lema='play' sint='V'>playing</word>

</sentence>

   </corpus>

•   Optional verbal list. A verbal list need to be a file UTF-8 like it:

  • verb1

  • verb2

  • verb3