def OnExtract(self, events):
text = self.sourcePage.GetValue().strip()
keyword_result = ''
result = ''
if text != '':
if self.languageType.GetSelection() == 1:
sentences_percent = self.sentencesPercent.GetValue()
similarity_function = self.similarityFunction.GetValue()
print similarity_function
extractor = Extractor(
stop_words_file='./TextRank/trainer/stopword_zh.data')
keyword, keyphrase = extractor.keyword_train(text=text)
abstract = extractor.sentence_train(
text,
sentences_percent=sentences_percent,
sim_func=similarity_function)
keyword_result = '/'.join(keyword)
keyword_result += '\n关键短语：\n' + '/'.join(keyphrase)
result += '。'.join(abstract) + r'。'
self.abstractPage.SetValue(result)
#设置文本样式
#f = wx.Font(10, wx.ROMAN, wx.NORMAL, wx.NORMAL, True) #创建一个字体
#self.abstractPage.SetStyle(0, len(result), wx.TextAttr('black',wx.NullColor,f))
self.keywordPage.SetValue(keyword_result)
else:
art_type = self.articleType.GetSelection()
extractor = EnExtractor(
stop_words_file='./TextRank/trainer/stopword_en.data')
if art_type == 1:
keyphrase = extractor.keyphrase_train(
text, article_type='Abstract')
keyword_result = 'Keyphrases:\n' + '/'.join(keyphrase)
else:
sentences_percent = self.sentencesPercent.GetValue()
similarity_function = self.similarityFunction.GetValue()
keyphrase = extractor.keyphrase_train(
text, article_type='Fulltext')
summary = extractor.summary_train(
text,
sentences_percent=sentences_percent,
sim_func=similarity_function)
keyword_result = '/'.join(keyphrase)
result += ' ' + ' '.join(summary)
self.abstractPage.SetValue(result)
#设置文本样式
f = wx.Font(10, wx.ROMAN, wx.NORMAL, wx.NORMAL, True) #创建一个字体
self.abstractPage.SetStyle(
0, len(result), wx.TextAttr('black', wx.NullColor, f))
self.keywordPage.SetValue(keyword_result)
else:
#test
#sentences_percent = self.sentencesPercent.GetValue()
#print filter(lambda x:x.isdigit(), sentences_percent)
print "No article"
def get_keyphrases(basepath, filelist):
correct_num = 0
system_key_num = 0
wrong_file = []
for f in filelist:
#print 'deal with:',f
evaluation = EnExtractor()
text = open(basepath+'/abstracts/'+f,'r').read()
system_keyphrases = evaluation.keyphrase_train(text=text, article_type='Abstract')
system_key_num += len(system_keyphrases)
#tag_word = evaluation.get_tag(text=text)
content = open(basepath+'/keywords/'+f,'r').read()
mannual_keyphrases = content.split(';')
'''test for 0 corret_num file'''
tmp = eval_keyphrase(system_keyphrases,mannual_keyphrases)
correct_num += tmp
print 'file-',f,':',correct_num
if tmp == 0:
wrong_file.append(f)
return correct_num ,system_key_num , wrong_file
def get_keyphrases(basepath, filelist):
correct_num = 0
system_key_num = 0
wrong_file = []
for f in filelist:
#print 'deal with:',f
evaluation = EnExtractor()
text = open(basepath + '/abstracts/' + f, 'r').read()
system_keyphrases = evaluation.keyphrase_train(text=text,
article_type='Abstract')
system_key_num += len(system_keyphrases)
#tag_word = evaluation.get_tag(text=text)
content = open(basepath + '/keywords/' + f, 'r').read()
mannual_keyphrases = content.split(';')
'''test for 0 corret_num file'''
tmp = eval_keyphrase(system_keyphrases, mannual_keyphrases)
correct_num += tmp
print 'file-', f, ':', correct_num
if tmp == 0:
wrong_file.append(f)
return correct_num, system_key_num, wrong_file
def get_summary(sentences):
extractor = EnExtractor()
summary = extractor.summary_train(sentences)
return summary
def get_summary(sentences): extractor = EnExtractor() summary = extractor.summary_train(sentences) return summary
