def test_document(self):
# Assert Document properties.
# Test with different input types.
for constructor, w in ((vector.Document, "The cats sat on the mat."),
(vector.Document,
["The", "cats", "sat", "on", "the",
"mat"]), (vector.Document, {
"cat": 1,
"sat": 1,
"mat": 1
}), (vector.Document,
Text(parse("The cats sat on the mat."))),
(vector.Document,
Sentence(parse("The cats sat on the mat."))),
(vector.Document.open, self.path1)):
# Test copy.
v = constructor(w,
stemmer=vector.LEMMA,
stopwords=False,
name="Cat",
type="CAT")
v = v.copy()
# Test properties.
self.assertEqual(v.name, "Cat")
self.assertEqual(v.type, "CAT")
self.assertEqual(v.count, 3)
self.assertEqual(v.terms, {"cat": 1, "sat": 1, "mat": 1})
# Test iterator decoration.
self.assertEqual(sorted(v.features), ["cat", "mat", "sat"])
self.assertEqual(sorted(v), ["cat", "mat", "sat"])
self.assertEqual(len(v), 3)
self.assertEqual(v["cat"], 1)
self.assertEqual("cat" in v, True)
print "pattern.vector.Document"
def analyze(self):
text = Text(self.clean_text)
for sentence in text:
print '-----SENTENCE' + str(sentence.string)
print sentence.subjects
print sentence.verbs
for word in sentence:
print word.string
print word.lemma
print word.type
print word.chunk
def convert_pattern_format(text):
"""
Text is parsed through pattern's parsing function into a standardized format.
"""
parsed_text = []
# parse text via Pattern's parser
pattern_parsed_text = Text(parse(text, relations=True, lemmata=True))
for sentence in pattern_parsed_text:
s = Sentence()
s.string = remove_blanks(sentence.string)
for word in sentence:
# Patterns tags for each word in the sentence are stored in a new Word-object
w = Word()
w.string = word.string
w.lemma = word.lemma
w.index = word.index
w.tag = word.type
w.entity = ""
# each word is appended to a Sentence-object
s.words.append(w)
# each Sentence-object is appended to an array
parsed_text.append(s)
return parsed_text
import os, sys
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", ".."))
from pattern.en import parse, Text
# The easiest way to analyze the output of the parser is to create a Text.
# A Text is a "parse tree" of linked Python objects.
# A Text is essentially a list of Sentence objects.
# Each Sentence is a list of Word objects.
# Each Word can be part of a Chunk object, accessible with Word.chunk.
s = "I eat pizza with a silver fork."
s = parse(s)
s = Text(s)
# You can also use the parsetree() function,
# which is the equivalent of Text(parse()).
print(s[0].words) # A list of all the words in the first sentence.
print(s[0].chunks) # A list of all the chunks in the first sentence.
print(s[0].chunks[-1].words)
print("")
for sentence in s:
for word in sentence:
print(word.string, word.type, word.chunk, word.pnp)
# A Text can be exported as an XML-string (among other).
print("")
print(s.xml)
def parse(self, text):
self.valid_sentences = []
self.faulty_sentences = []
self.statistic.properties['valid_sentences'] = 0
self.statistic.properties['too_few_words'] = 0
self.statistic.properties['first_word_is_number'] = 0
self.statistic.properties['sentence_contains_brackets'] = 0
self.statistic.properties['sentence_contains_number'] = 0
self.statistic.properties['sentence_too_many_comma'] = 0
self.statistic.properties['too_many_short_words'] = 0
self.statistic.properties['sentence_in_german'] = 0
self.statistic.properties['sentence_in_french'] = 0
self.statistic.properties['sentence_in_spanish'] = 0
self.statistic.properties['sentence_in_italian'] = 0
self.statistic.properties['sentence_in_dutch'] = 0
self.statistic.properties['sentence_not_english'] = 0
self.statistic.properties['begins_with_punctuation'] = 0
self.statistic.properties['weird_chars'] = 0
self.statistic.properties['first_not_upper'] = 0
self.statistic.properties['last_not_dot'] = 0
self.statistic.properties['too_many_dots'] = 0
text = self.replace(text)
text = Text(
parse(text,
tokenize=True,
tags=True,
chunks=True,
relations=False,
lemmata=False,
encoding='utf-8',
tagset=None))
for sentence in text:
replaced_string = self.replace(sentence.string)
if len(sentence.words) < self.MIN_WORD_COUNT:
# too few words in the sentence
# removes sentences like these: https://gist.github.com/mrzl/32b9763bd943c18cb77cd1167a87640a
self.statistic.properties['too_few_words'] += 1
self.faulty_sentences.append(sentence)
continue
if self.get_perc_single_char_words(sentence) > 0.5:
# too many short words
self.statistic.properties['too_many_short_words'] += 1
self.faulty_sentences.append(sentence)
continue
if replaced_string[0].isdigit():
# first word of the sentence is a number
self.statistic.properties['first_word_is_number'] += 1
self.faulty_sentences.append(sentence)
continue
if self.contains_tag(sentence, "(", 3) \
or self.contains_tag(sentence, ")", 3):
# the sentence contains either ( )
self.statistic.properties['sentence_contains_brackets'] += 1
self.faulty_sentences.append(sentence)
continue
if self.get_count_tag(sentence, "CD") > 3:
# the sentence contains a number
self.statistic.properties['sentence_contains_number'] += 1
self.faulty_sentences.append(sentence)
continue
if self.get_count_tag(sentence, ",") > 6:
# the sentence has more than 2 occurrences of a comma(,)
self.statistic.properties['sentence_too_many_comma'] += 1
self.faulty_sentences.append(sentence)
continue
if replaced_string[0] in [
u'.', u'\'', u';', u'~', u':', u'-', u'·', u'‘', u'’',
u'\\'
]:
# sentence begins with punctuations
self.statistic.properties['begins_with_punctuation'] += 1
self.faulty_sentences.append(sentence)
continue
if self.get_count_of_special_chars(sentence) > 3:
# sentence contains weirdly escaped chars
self.statistic.properties['weird_chars'] += 1
self.faulty_sentences.append(sentence)
continue
if not replaced_string[0].isupper():
# first char is not upper case
self.statistic.properties['first_not_upper'] += 1
self.faulty_sentences.append(sentence)
continue
if not sentence.string[-1] in ['.', '?', '!']:
# last char not a dot
self.statistic.properties['last_not_dot'] += 1
self.faulty_sentences.append(sentence)
continue
if sentence.string.count('.') > 3:
# too many dots in the sentence
self.statistic.properties['too_many_dots'] += 1
self.faulty_sentences.append(sentence)
continue
try:
classified = langid.classify(sentence.string)[0]
if classified in 'en':
self.valid_sentences.append(sentence)
continue
elif classified in 'de':
self.statistic.properties['sentence_in_german'] += 1
self.faulty_sentences.append(sentence)
continue
elif classified in 'fr':
self.statistic.properties['sentence_in_french'] += 1
self.faulty_sentences.append(sentence)
continue
elif classified in 'es':
self.statistic.properties['sentence_in_spanish'] += 1
self.faulty_sentences.append(sentence)
continue
elif classified in 'nl':
self.statistic.properties['sentence_in_dutch'] += 1
self.faulty_sentences.append(sentence)
continue
elif classified in 'it':
self.statistic.properties['sentence_in_italian'] += 1
self.faulty_sentences.append(sentence)
continue
else:
self.statistic.properties['sentence_not_english'] += 1
self.faulty_sentences.append(sentence)
continue
except:
self.statistic.properties['sentence_not_english'] += 1
self.faulty_sentences.append(sentence)
self.statistic.properties['valid_sentences'] += len(
self.valid_sentences)
print('Parsed ' + str(len(self.valid_sentences)) +
' proper sentences.')
sum = 0
for key, value in self.statistic.properties.items():
if key not in 'valid_sentences':
sum += value
print('Discarded ' + str(sum) + ' invalid sentences.')
tokenize=True, # Split punctuation marks from words?
tags=True, # Parse part-of-speech tags? (NN, JJ, ...)
chunks=True, # Parse chunks? (NP, VP, PNP, ...)
relations=False, # Parse chunk relations? (-SBJ, -OBJ, ...)
lemmata=False, # Parse lemmata? (ate => eat)
encoding='utf-8', # Input string encoding.
tagset=None) # Penn Treebank II (default) or UNIVERSAL.
print repr(s)
for sentence in s:
for chunk in sentence.chunks:
print chunk.type, [(w.string, w.type) for w in chunk.words]
for sentence in tree(open('data/input/tagged.txt'),
token=[WORD, POS, CHUNK]): # CHECK FOR ERROR
print sentence
# text
text = Text(open('data/input/corpus.txt'),
token=[WORD, POS, CHUNK, PNP, REL, LEMMA])
# text = Text.from_xml('data/input/multilingual-all-words.en.xml') # Reads an XML string generated with Text.xml.
print text.string # 'The cat sat on the mat .'
print text.sentences # [Sentence('The cat sat on the mat .')]
print text.copy()
print text.xml
# sentence
# sentence = Sentence(open('data/input/corpus.txt'), token=[WORD, POS, CHUNK, PNP, REL, LEMMA])
# sentence = Sentence.from_xml(xml)
print sentence.parent # Sentence parent, or None.
print sentence.id # Unique id for each sentence.
print sentence.start # 0
print sentence.stop # len(Sentence).
print sentence.string # Tokenized string, without tags.
print sentence.words # List of Word objects.
print sentence.lemmata # List of word lemmata.
def __init__(self,data):
# Don't convert to lower string as CAPS are important too. But Implement lower
if isinstance(data,basestring):
self.DATA = data
self.PARSE = parse(self.DATA)
self.DOC = Text(self.PARSE)
