David S. Batista davidsbatista

Follow the tutorial to install Tensorflow implementation of Syntaxet
Download the trained Portuguese model from universal dependencies and unzipped it.
Run the following command on the shell

MODEL_DIRECTORY=/Users/dbatista/Downloads/Portuguese
cd /Users/dbatista/models/syntaxnet
echo "Quase 900 funcionários do Departamento de Estado assinaram memorando que critica Trump" | syntaxnet/models/parsey_universal/parse.sh $MODEL_DIRECTORY

	## Full query gives timeout

	SELECT ?entity ?entityLabel
	WHERE
	{
	?item_type wdt:P279 wd:Q1269299 . # Item's type is a subclass of type of business entity
	?business_entity wdt:P31 ?item_type .
	?entity rdfs:label ?entityLabel .
	?entity skos:altLabel ?alternativeNames .
	FILTER (lang(?alternativeNames) = "de")

	#defaultView:Map
	SELECT DISTINCT ?companyLabel ?company ?headquarters ?headquartersLabel ?coordinates WHERE {
	VALUES ?stock_markets {wd:Q155718 wd:Q378967 wd:Q595622 wd:Q874430} .
	?company wdt:P361 ?stock_markets .
	?company rdfs:label ?companyLabel . FILTER(LANG(?companyLabel) = "en")
	?company wdt:P159 ?headquarters.
	?headquarters rdfs:label ?headquartersLabel . FILTER(LANG(?headquartersLabel) = "en")
	?headquarters wdt:P625 ?coordinates
	}
	ORDER BY ASC(?companyLabel)

	#defaultView:Map
	SELECT DISTINCT ?dam ?name ?elevation ?coords WHERE {
	# ?dam wdt:P31/wdt:P279* wd:Q12323 .
	?dam wdt:P31/wdt:P279* wd:Q15911738 . # only hidroelectric cpower
	?dam wdt:P17 wd:Q45 . # only located in Portugal
	?dam rdfs:label ?name FILTER(LANG(?name) = "pt") .
	?dam p:P2044 ?elevation_sea_level .
	?elevation_sea_level ps:P2044 ?elevation .
	?dam p:P625 ?coordinataes .
	?coordinataes ps:P625 ?coords;

	#defaultView:Map
	SELECT DISTINCT ?dam ?coords ?lat ?long WHERE {
	?dam wdt:P31/wdt:P279* wd:Q12323 .
	?dam wdt:P17 wd:Q45 .
	?dam p:P625 ?coordinataes .
	?coordinataes ps:P625 ?coords;
	psv:P625 [
	wikibase:geoLatitude ?lat;
	wikibase:geoLongitude ?long
	] .

	SELECT ?iata_code ?airport_name (COUNT(?connectsairport) AS ?nr_connections)
	WHERE
	{
	VALUES ?airport { wd:Q17480 wd:Q9694 wd:Q160556 wd:Q403671}
	?airport wdt:P238 ?iata_code.
	?airport rdfs:label ?airport_name.
	OPTIONAL {
	?airport wdt:P81 ?connectsairport.
	}
	FILTER(LANG(?airport_name) = "en")

	SELECT DISTINCT ?capital ?capital_label ?pop ?code WHERE {
	?county wdt:P31/wdt:P279* wd:Q47168 .
	?county wdt:P36 ?capital .
	?capital rdfs:label ?capital_label .
	?capital wdt:P1082 ?pop .
	?county wdt:P131 ?state .
	?state wdt:P31 wd:Q35657 .
	?state wdt:P300 ?code .
	FILTER(LANG(?capital_label) = "en")
	FILTER(?pop >= 50000).

	SELECT DISTINCT ?station_label ?ibnr_stmt ?x ?ibnr_type_stmnt ?label WHERE
	{
	wd:Q376589 rdfs:label ?station_label.
	wd:Q376589 p:P954 ?ibnr_stmt.
	?ibnr_stmt ps:P954 ?x.
	?ibnr_stmt pq:P518 ?ibnr_type_stmnt.
	?ibnr_type_stmnt rdfs:label ?label.
	FILTER(lang(?station_label) = 'de')
	FILTER(lang(?label) = 'de')
	}

	SELECT ?iata_code ?places_served_label(LANG(?places_served_label) AS ?places_served_label_lang) ?country_code
	WHERE {
	?item wdt:P238 ?iata_code.
	?item wdt:P931 ?places_served.
	?places_served rdfs:label ?places_served_label.
	?places_served wdt:P17 ?country.
	?country wdt:P297 ?country_code.
	FILTER( LANG(?places_served_label) = "de" \|\| LANG(?places_served_label) = "en").
	}

	import numpy as np
	import matplotlib.pyplot as plt

	from sklearn import svm, datasets
	from sklearn.model_selection import train_test_split
	from sklearn.metrics import confusion_matrix
	from sklearn.utils.multiclass import unique_labels

	def plot_confusion_matrix(y_true, y_pred, classes, normalize=False, title=None, cmap=plt.cm.Blues):
	"""