arademaker

> tmp.1 <- data[grep("person",data$lex..file2,invert=TRUE),]
> subset(tmp.1, rel == "nomlex:agent", c(1,3,4,5,6,7,8,9))
           word1         word2    synset1    synset2       lex..file1     lex..file2 suffix      class
77       adaptar     adaptador 00299580-v 02678897-n      verb.change  noun.artifact     or    REGULAR
103     fabricar    fabricante 01653442-v 08060446-n    verb.creation     noun.group    nte    REGULAR
237     tripular    tripulação 01089614-v 08273167-n verb.competition     noun.group    ção    REGULAR
254      imigrar     imigração 00413432-v 08413834-n      verb.change     noun.group    ção    REGULAR
273   transferir transferência 02220461-v 01107932-n  verb.possession       noun.act    cia    REGULAR
294       servir       serviço 02541251-v 08186047-n      verb.social     noun.group      o REGRESSIVE
357         roer        roedor 01445597-v 02329401-n     verb.contact    noun.animal     or    REGULAR

arademaker

word	sense	sense
cross	wordsense-01913997-v-1	wordsense-02023396-v-2
farsightedness	wordsense-14555214-n-2	wordsense-14554011-n-4
get_up	wordsense-00018405-v-1	wordsense-01974062-v-4
lift	wordsense-01455184-v-2	wordsense-01974062-v-2
raise	wordsense-01975587-v-1	wordsense-01974062-v-1
make	wordsense-01646075-v-1	wordsense-01645601-v-3
do	wordsense-02561995-v-1	wordsense-01645601-v-2

arademaker

$ awk '{print $2, $2, $4, "#"}' train.pt | /media/data/src/freeling/src/utilities/train-tagger/bin/change-corpus-dict pt /usr/local/share/freeling/
O O DET # o DA0MS0 0.946534 o PD0MS000 0.0337339 o PP3MSA00 0.0197186 o NCMS000 1.38766e-05
objetivo objetivo NOUN # objetivo AQ0MS0 0.554567 objetivo NCMS000 0.390889 objetivo NP00000 0.0545441
dos dos ADP # de SPS00 1
principais principais ADJ # principal AQ0CP0 0.99061 principais NCMP000 0.00469484 principal NCMP000 0.00469484
hotéis hotéis NOUN # hotel NCMP000 1
da da ADP # de SPS00 1
cidade cidade NOUN # cidade NCFS000 1
é é VERB # ser VMIP3S0 1
que que CONJ # que PR0CN000 0.586684 que CS 0.39617 que PT0CN000 0.0167849 que PE0CN000 0.000361319

arademaker

$ java -jar /usr/local/HermiT/HermiT.jar --classify test-complet.ttl
Exception in thread "main" org.semanticweb.owlapi.reasoner.InconsistentOntologyException: Inconsistent ontology
	at org.semanticweb.HermiT.Reasoner.throwInconsistentOntologyExceptionIfNecessary(Unknown Source)
	at org.semanticweb.HermiT.Reasoner.checkPreConditions(Unknown Source)
	at org.semanticweb.HermiT.Reasoner.precomputeInferences(Unknown Source)
	at org.semanticweb.HermiT.cli.CommandLine$ClassifyAction.run(Unknown Source)
	at org.semanticweb.HermiT.cli.CommandLine.main(Unknown Source)

arademaker

stardog-admin server start

stardog-admin db drop wn
stardog-admin db create -n wn wn30.ttl

stardog data add wn test.ttl

stardog-admin icv add -v wn wn30.ttl
stardog icv validate -v -r wn   ; the r option is necessary

arademaker

Portuguese POS tagged corpora

Dados CoNLL-X

Este primeiro arquivo foi obtido do CoNLL-X Shared Task. É o arquivo de teste disponibilizado. A quarta coluna, V4, contém a tag.

arademaker

;; utilities

(defun combinations (&rest lists)
	   (if (car lists)
	       (mapcan (lambda (inner-val)
			 (mapcar (lambda (outer-val)
				   (cons outer-val inner-val))
				 (car lists)))
		       (apply #'combinations (cdr lists)))
	       (list nil)))

arademaker

<?xml version="1.0" encoding="ISO-8859-1"?>

<!ELEMENT EM (#PCDATA) >
<!ATTLIST EM
	  ID      CDATA #REQUIRED 
	  CATEG   CDATA #IMPLIED
	  TIPO    CDATA #IMPLIED
	  COMENT  CDATA #IMPLIED
	  SUBTIPO CDATA #IMPLIED >

arademaker

-- testing double without type information
double x = x + x

-- testing the first example
qsort :: Ord a => [a] -> [a]
qsort []     = []
qsort (p:xs) = (qsort lesser) ++ [p] ++ (qsort greater)
  where
    lesser  = filter (< p) xs
    greater = filter (>= p) xs

arademaker

#!/opt/local/bin/python3.2
# -*- encoding: utf-8 -*-

import sys
import json

sys.path.append('.../myfreeling/APIs/python')
import freeling
import codecs
import re