def parseTextToSentences(text):
punkt_param = PunktParameters()
punkt_param.abbrev_types = set(['dr', 'vs', 'mr', 'ms', 'mrs', 'prof', 'inc', 'no', 'e.g', 'i.e'])
sentence_splitter = PunktSentenceTokenizer(punkt_param)
data = text
data = data.replace('?"', '? "').replace('!"', '! "').replace('."', '. "')
sentences = []
for para in data.split('\n'):
if para:
sentences.extend(sentence_splitter.tokenize(para))
return sentences
def removeTag(data):
cleanData = ''
for word in data.split():
if '<' and '>' in word:
word = '. '
cleanData = cleanData + word + ' '
return cleanData
def process_tokenized(data):
"""Process tokenized data and remove the newlines."""
newdata = ""
data = data.split("\n")
for line in data:
if LINE_SEPARATOR not in line:
newdata = "{}{}".format(newdata, line.replace("\n", ""))
else:
newdata = "\n\n{}\n{}\n".format(newdata, line)
return newdata
def gloss_count_ratio(data, get_ratio=0):
term_count = 0
word_count = 0
for key, values in allkeys.iteritems():
if key in data:
term_count += 1
if (get_ratio):
word_count += len(data.split())
return term_count, float(term_count) / word_count
else:
return term_count
def parse_text_to_sentences(text):
punkt_param = PunktParameters()
punkt_param.abbrev_types = set(
['dr', 'vs', 'mr', 'ms', 'mrs', 'prof', 'inc', 'no', 'e.g', 'i.e'])
sentence_splitter = PunktSentenceTokenizer(punkt_param)
data = text
data = data.replace('?"', '? "').replace('!"', '! "').replace('."', '. "')
sentences = []
for para in data.split('\n'):
if para:
sentences.extend(sentence_splitter.tokenize(para))
return sentences
def extract_labels(self, binary=True):
"""
Loads labeled dataframe and extract list of labeled subtitle ids.
"""
data = self.key_to_labels.get_contents_as_string()
valid_ratings = self.context.broadcast([u'R', u'PG-13', u'PG', u'G', u'NC-17'])
labels_rdd = self.context.parallelize(data.split('\n')) \
.filter(lambda line: line != '' and not 'IDSubtitle' in line) \
.map(lambda line: (line.split(',')[0], line.split(',')[1])) \
.filter(lambda (file_id, rating): rating in valid_ratings.value)
if binary:
labels_rdd = labels_rdd.map(lambda (file_id, rating): (file_id, (rating != 'R')*'NOT_' + 'R'))
return labels_rdd.sortByKey().cache() # for lookups
def gloss_count_ratio(data, get_ratio=0):
term_count = 0
word_count = 0
for key, values in allkeys.iteritems():
if key in data:
#Found term. So increase its count
recalldict[key] += 1
term_count += 1
if (get_ratio):
word_count += len(data.split())
#print "{},{},{}".format(float(term_count)/word_count,term_count,word_count)
return float(term_count) / word_count
else:
return term_count
def prepare_text(author=None, data=None):
# text_column = "text"
# author_column = "author"
# train_file = os.path.join(os.getcwd(), "my_train.csv")
# test_file = os.path.join(os.getcwd(), "my_test.csv")
# sentences = tokenize_text(data)
sentences = data.split(" ")
train_sentences = " ".join(sentences[:int(len(sentences) * 0.50)])
test_sentences = " ".join(sentences[int(len(sentences) * 0.50):])
train_res = [author] * 2
test_res = [author] * 2
return train_sentences, test_sentences, train_res, test_res
def handleonetxtfile(inpname):
basename = os.path.basename(inpname).split('.')[0]
basename_txt = os.path.join(basename + ".txt")
outname = os.path.join(basename + "_setnence_converted.txt")
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
fp = open(basename_txt)
data = fp.read()
paragraphs = [p for p in data.split('\n') if p]
with open(outname, "a") as out:
for paragraph in paragraphs:
out.write('\n')
out.write('\n'.join(tokenizer.tokenize(paragraph)))
out.write('\n')
out.close()
def splitIntoSentences2(file_name):
punkt_param = PunktParameters()
punkt_param.abbrev_types = set(['dr', 'vs', 'mr', 'mrs', 'prof', 'inc'])
sentence_splitter = PunktSentenceTokenizer(punkt_param)
fp = open(file_name)
data = fp.read()
data = data.replace('?"', '? "').replace('!"', '! "').replace('."', '. "')
sentences = []
for para in data.split('\n'):
if para:
sentences.extend(sentence_splitter.tokenize(para))
# print '\n-----\n'.join(sentences)
return sentences
def Tokenization(data, concept, stem, removeStopwords):
if concept == False:
data = BeautifulSoup(data).get_text()
data = re.sub("\r\n", " ", data)
data = re.sub("[^a-zA-Z0-9_]", " ", data)
data = data.lower()
if stem == True:
stemmer = PorterStemmer()
data = stemmer.stem(data)
words = data.split()
if removeStopwords == True:
stops = set(stopwords.words("english"))
words = [w for w in words if not w in stops]
return words
def split_sentences_nltk(
file): #returns the topcs separated into sentenses and topic titles
try:
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
except:
nltk.download('punkt') # needed only for the first time
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
data = file.read()
data_1 = data.split("changeTopicHere.")
topics, data_1 = find_topic(data_1)
topics_data = []
i = 0
for x in xrange(0, len(data_1)):
topics_data.append("\n".join(tokenizer.tokenize(data_1[x])))
return topics, topics_data
def openparsefile(self):
self.clear()
self.filename = tkFileDialog.askopenfilename(**self.file_opt)
if not self.filename:
return
data = open(self.filename, 'r').read()
self.initree = {}
filtdata = ""
self.ininame = {}
for entry in data.split('\n\n'):
if not entry:
continue
for line in entry.split("\n"):
elements = line.split()
if len(elements) < 4:
continue
if len(elements) == 5:
lex_list = word_to_features(elements[0])
fset = lex_search(lex_list, {}, self.alltrees)
for cata in fset:
if elements[4] in fset[cata]:
tree = fset[cata][elements[4]]
tree.lexicalize()
tree._lex = False
self.initree[elements[0]] = tree
self.ininame[elements[0]] = elements[4]
self.relmap[elements[0]] = elements[1]
self.tagmap[elements[0]] = elements[3]
filtdata += "\t".join(elements[:4])
filtdata += '\n'
filtdata += "\n\n"
self.graphs = [
DependencyGraph(entry) for entry in filtdata.split('\n\n')[:-1]
if entry
]
trees = [graph.tree() for graph in self.graphs]
count = 0
for t in trees:
key = 'tree ' + str(count)
self._import_trees[key] = t
count += 1
self.populate_tree('', self._import_trees)
self._import_trees_selected = True
self.init_selector(trees[0])
def openparsefile(self):
self.clear()
self.filename = tkFileDialog.askopenfilename(**self.file_opt)
if not self.filename:
return
data = open(self.filename, 'r').read()
self.initree = {}
filtdata = ""
self.ininame = {}
for entry in data.split('\n\n'):
if not entry:
continue
for line in entry.split("\n"):
elements = line.split()
if len(elements) < 4:
continue
if len(elements) == 5:
lex_list = word_to_features(elements[0])
fset = lex_search(lex_list, {}, self.alltrees)
for cata in fset:
if elements[4] in fset[cata]:
tree = fset[cata][elements[4]]
tree.lexicalize()
tree._lex = False
self.initree[elements[0]] = tree
self.ininame[elements[0]] = elements[4]
self.relmap[elements[0]] = elements[1]
self.tagmap[elements[0]] = elements[3]
filtdata += "\t".join(elements[:4])
filtdata += '\n'
filtdata += "\n\n"
self.graphs = [DependencyGraph(entry)
for entry in filtdata.split('\n\n')[:-1] if entry]
trees = [graph.tree() for graph in self.graphs]
count = 0
for t in trees:
key = 'tree '+ str(count)
self._import_trees[key] = t
count += 1
self.populate_tree('', self._import_trees)
self._import_trees_selected = True
self.init_selector(trees[0])
date = sys.argv[2]
article = 1
total_avg_polarity = 0
while (article <= 5):
file = stock + '/' + date + '_2016/' + str(article) + '.txt'
try:
fp = open(file)
article += 1
data = fp.read()
# print sentences
title = data.split('\n', 1)[0]
content = data
# Create a SummaryTool object
st = SummaryTool()
# Build the sentences dictionary
sentences_dic = st.get_sentences_ranks(content)
# Build the summary with the sentences dictionary
sentences = st.get_summary(title, content, sentences_dic)
# print sentences
## instantiate senticnet
root = inp.getroot()
candidates = []
for child in root.iter():
if child.tag == "P":
child = tokenizer.tokenize(child.text)
for i in child:
candidates.append(i)
data = "Discuss conditions on American Indian reservations or among Native American communities. Include the benefits and drawbacks of the reservation system. Include legal privileges and problems."
print "Query Sentence"
print data
print "\n"
data = data.split(" ")
queryRel = []
for word in data:
for i,j in enumerate(wn.synsets(word)):
for l in j.lemmas():
queryRel.append(l.name())
#queryRel.append(l.lemma_names() for l in j.hypernyms())
for l in j.hypernyms():
for k in l.lemma_names():
queryRel.append(k)
for l in j.hyponyms():
for k in l.lemma_names():
queryRel.append(k)
def LLR(e):
def file_to_words(self):
data = None
with open(self.corpus_file) as corpus:
data = corpus.read()
words = data.split()
return words
def get_raw_paragraph(
fileid
): #TODO test if this works with yahoo! corpus as well (encoding might differ)
data = corpus.raw(fileid)
return data.split(u"\r\n \r\n")
# print(x)
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
data = f.read()
result = ('\n------\n'.join(tokenizer.tokenize(data)))
# print(result)
# data = f.read()
# for e in firstprocess:
# # print(e)
# # print(e.index('\n'))
# s = e.split()
# print(s)
# break
allSentences = []
data2 = data.split()
newWord = False
upperCaseWord = False
lowerCaseWord = False
currentSentence = ""
for e in data2:
if e.isupper():
upperCaseWord = True
if lowerCaseWord is True:
newWord = True
allSentences.append(currentSentence)
currentSentence = currentSentence + ' ' + e
if e.islower():
if upperCaseWord is True:
newWord = True
import nltk.data
import nltk
import language_check
import string
import re
tool = language_check.LanguageTool('en-US')
nltk.download('punkt')
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
lines = []
fp = open("./data/dbt_handouts_pass1.txt")
data = fp.read()
data = data.replace("\n", ' ').replace('•• ',
'').replace('*',
'').replace('Other:', '')
data = re.sub(r'\d+', ' ', data)
data = re.sub(r'[^\x00-\x7f]', r'', data)
data = ' '.join(data.split())
for x in tokenizer.tokenize(data):
matches = tool.check(x)
words = x.split(' ')
if len(matches) == 0 and len(words) >= 8:
lines.append(x)
print('\n'.join(lines))
f = open("./data/dbt_handouts_pass2.txt", "w")
f.write('\n'.join(tokenizer.tokenize(data)))
f.close()
num=0
for fileName in list_of_files:
s=[]
p=[]
r=[]
t=[]
q=[]
a=[]
fin=open(fileName,"r")
data=fin.read()
print(fileName,"check3")
print(data)
#inp=input("enter your opinion:")
inp=input("press 1 for sports 2 for education 3 for entertainment 4 for politics 5 for business")
print(inp)
for line in data.split('\n'):
#print(line)
r.append(line)
#print(type(s))
s.append(word_tokenize(line))
#print(word_tokenize(line))
fin.close() #closes file
#print(s)
for i in range(len(s)):
r.append(i)
#print(r)
t = word_tokenize(str(r))
print(t)
p = nltk.corpus.stopwords.words('english')
#print(p)
continue
except:
continue
try:
if (len(note.get_text().split())) == 1:
footNote = note.attrs['n'] + ' ' + note.get_text()
dic[note.attrs['n']] = note.get_text()
if (len(note.get_text().split())) > 1:
footNote = note.attrs['n'] + ' ' + note.get_text()
for word in footNote.split():
#if word[:5] == 'https':
if 'htt' in word:
data = data + word + ' '
if word.isdigit():
data = data + word + ' '
for multi in data.split():
if multi.isdigit():
key = multi
continue
dic[key] = multi
listXMLfile.append(filenameXML)
outputCSV_writer.writerow([filenameXML, dic[key]])
print(count)
except:
continue
#print(dic)
'''
#// extracting the sentance
cleanSentence = None
sentences = split_into_sentences(contents)
cmu = cmudict.dict()
# webster = loadWebster(webster)
################## open file ##################
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
fp = open("sample.txt")
# fp = open("collier.txt")
# fp = open("frost_woods.txt")
# fp = open("pope_windsor_forest.txt")
# fp = open("paradise_lost.txt")
# fp = open("sonnet_xv.txt")
# fp = open("thomson_seasons.txt")
data = fp.read()
data = data.split('\n') ## line breaking.
# exclude = set(string.punctuation)
exclude = set('!"#$%&()*+,./:;<=>?@[\\]^_`{|}~')
# exclude all string.punctuation except apostrophe (I think?)
# and hyphen (29 may 2013)
# note that i remove all of string.punctuation at the end of
# the replED function
lines = [] #to store the poem
for datum in data:
'''
This is really ugly, but, I needed to replace -'d endings
with -ed endings, and the only place I could think of doing
it was when first creating the lines.
As this loop starts, apostrophes should be the only punctuation
def file_to_words(self):
data = None
with open(self.corpus_file) as corpus:
data = corpus.read()
words = data.split()
return words
