def hqd_word_segment(read_directory, write_directory1, write_directory2, write_directory3, write_directory4, write_directory5):
'''
:param read_directory:
:param write_directory1:
:param write_directory2:
:param write_directory3:
:param write_directory4:
:param write_directory5:
'''
stopwords_list1 = get_stopwords1()
file_number = sum([len(files) for root, dirs, files in os.walk(read_directory)])
for i in range(file_number):
time_series = []
class_tag = []
content_with_tag = []
content_without_tag = []
all_weibo_word = []
f = open(read_directory + '/' + str(i + 1) + '.txt', 'rb')
line = f.readline()
while line:
this_line = line.strip().split('\t')
#this_id = this_line[0]
this_time = time.mktime(time.strptime(this_line[2], '%Y/%m/%d %H:%M'))
time_series.append(str(this_time))
class_tag.append(this_line[5])
try:
this_text = this_line[6]
except:
this_text = " "
wd_with_tag = word_segment(this_text, stopwords_list1)
wd_without_tag = [x.split('/')[0] for x in wd_with_tag]
# 此处的词汇带有词性标注
for word in set(wd_with_tag).difference(all_weibo_word):
if word not in all_weibo_word:
all_weibo_word.append(word)
content_with_tag.append(" ".join(wd_with_tag))
content_without_tag.append(" ".join(wd_without_tag))
line = f.readline()
f.close()
quick_write_list_to_text(time_series, write_directory1 + '/' + str(i + 1) + '.txt')
quick_write_list_to_text(content_with_tag, write_directory2 + '/' + str(i + 1) + '.txt')
quick_write_list_to_text(content_without_tag, write_directory3 + '/' + str(i + 1) + '.txt')
quick_write_list_to_text(class_tag, write_directory4 + '/' + str(i + 1) + '.txt')
quick_write_list_to_text(all_weibo_word, write_directory5 + '/' + str(i + 1) + '.txt')
print "Segment %d Completed." % (i + 1)
def global_segment(read_filename, write_directory):
'''
所有微博文本分词，并获取微博的各项信息
:param read_filename:
:param write_directory:
'''
stopwords_list1 = get_stopwords1()
stopwords_list2 = get_stopwords2()
global_id = []
global_time = []
global_tag = []
weibo_sheet = open_sheet(read_filename)
weibo_row = weibo_sheet.nrows
print 'Number of the Weibo row: %d' % weibo_row
f1 = open(write_directory + '/weibo_content.txt', 'w')
f2 = open(write_directory + '/weibo_content2.txt', 'w')
for j in range(1, weibo_row):
weibo_id = str(int(weibo_sheet.cell(j, 0).value))
weibo_time = weibo_sheet.cell(j, 2).value
weibo_time = time_convert(weibo_time)
weibo_tag = str(int(weibo_sheet.cell(j, 5).value))
global_id.append(weibo_id)
global_time.append(str(weibo_time))
weibo_content = str(weibo_sheet.cell(j, 6).value)
fenci_result = word_segment(weibo_content, stopwords_list1, stopwords_list2)
f1.write(" ".join(fenci_result))
f1.write("\n")
fenci_without_tag = [x.split('/')[0] for x in fenci_result]
f2.write(" ".join(fenci_without_tag))
f2.write("\n")
global_tag.append(weibo_tag)
f1.close()
f2.close()
quick_write_list_to_text(global_id, write_directory + '/weibo_id.txt')
quick_write_list_to_text(global_time, write_directory + '/weibo_time.txt')
quick_write_list_to_text(global_tag, write_directory + '/weibo_class_tag.txt')
def microblog_extract(read_filename, write_directory):
'''
选出前8000条微博作为高质量文本并分词，获取微博的各项信息
:param read_filename:
:param write_directory:
'''
#file_number = sum([len(files) for root, dirs, files in os.walk(read_directory)])
select_number = 8000
stopwords_list1 = get_stopwords1()
stopwords_list2 = get_stopwords2()
#for i in range(file_number):
high_quality_weibo = []
high_quality_weibo2 = []
high_quality_id = []
high_quality_time = []
high_quality_tag = []
all_weibo_word = []
weibo_sheet = open_sheet(read_filename)
weibo_row = weibo_sheet.nrows
print 'Number of the Weibo row: %d' % weibo_row
count = 0
for j in range(1, weibo_row):
weibo_content = str(weibo_sheet.cell(j, 6).value)
fenci_result = word_segment(weibo_content, stopwords_list1, stopwords_list2)
if len(fenci_result) > 5 and (" ".join(fenci_result) not in high_quality_weibo):
count = count + 1
weibo_id = str(int(weibo_sheet.cell(j, 0).value))
weibo_time = weibo_sheet.cell(j, 2).value
weibo_time = time_convert(weibo_time)
weibo_tag = str(int(weibo_sheet.cell(j, 5).value))
high_quality_id.append(weibo_id)
high_quality_time.append(weibo_time)
high_quality_weibo.append(" ".join(fenci_result))
fenci_without_tag = [x.split('/')[0] for x in fenci_result]
high_quality_weibo2.append(" ".join(fenci_without_tag))
high_quality_tag.append(weibo_tag)
for word in set(fenci_result).difference(all_weibo_word):
if word not in all_weibo_word:
all_weibo_word.append(word)
if count >= select_number:
break
print len(high_quality_weibo)
# 按时间排序
itw = zip(high_quality_id, high_quality_time, high_quality_weibo, high_quality_weibo2, high_quality_tag)
itw1 = sorted(itw, key = itemgetter(1))
high_quality_weibo = []
high_quality_weibo2 = []
high_quality_id = []
high_quality_time = []
high_quality_tag = []
for each in itw1:
high_quality_id.append(each[0])
high_quality_time.append(str(each[1]))
high_quality_weibo.append(each[2])
high_quality_weibo2.append(each[3])
high_quality_tag.append(each[4])
quick_write_list_to_text(high_quality_id, write_directory + '/weibo_id.txt')
quick_write_list_to_text(high_quality_time, write_directory + '/weibo_time.txt')
quick_write_list_to_text(high_quality_weibo, write_directory + '/weibo_content.txt')
quick_write_list_to_text(high_quality_weibo2, write_directory + '/weibo_content2.txt')
quick_write_list_to_text(high_quality_tag, write_directory + '/weibo_class_tag.txt')
quick_write_list_to_text(all_weibo_word, write_directory + '/weibo_word.txt')
def microblog_extract(read_filename, write_directory):
'''
选出前8000条微博作为高质量文本并分词，获取微博的各项信息
:param read_filename:
:param write_directory:
'''
#file_number = sum([len(files) for root, dirs, files in os.walk(read_directory)])
select_number = 8000
stopwords_list1 = get_stopwords1()
stopwords_list2 = get_stopwords2()
#for i in range(file_number):
high_quality_weibo = []
high_quality_weibo2 = []
high_quality_id = []
high_quality_time = []
high_quality_tag = []
all_weibo_word = []
weibo_sheet = open_sheet(read_filename)
weibo_row = weibo_sheet.nrows
print 'Number of the Weibo row: %d' % weibo_row
count = 0
for j in range(1, weibo_row):
weibo_content = str(weibo_sheet.cell(j, 6).value)
fenci_result = word_segment(weibo_content, stopwords_list1,
stopwords_list2)
if len(fenci_result) > 5 and (" ".join(fenci_result)
not in high_quality_weibo):
count = count + 1
weibo_id = str(int(weibo_sheet.cell(j, 0).value))
weibo_time = weibo_sheet.cell(j, 2).value
weibo_time = time_convert(weibo_time)
weibo_tag = str(int(weibo_sheet.cell(j, 5).value))
high_quality_id.append(weibo_id)
high_quality_time.append(weibo_time)
high_quality_weibo.append(" ".join(fenci_result))
fenci_without_tag = [x.split('/')[0] for x in fenci_result]
high_quality_weibo2.append(" ".join(fenci_without_tag))
high_quality_tag.append(weibo_tag)
for word in set(fenci_result).difference(all_weibo_word):
if word not in all_weibo_word:
all_weibo_word.append(word)
if count >= select_number:
break
print len(high_quality_weibo)
# 按时间排序
itw = zip(high_quality_id, high_quality_time, high_quality_weibo,
high_quality_weibo2, high_quality_tag)
itw1 = sorted(itw, key=itemgetter(1))
high_quality_weibo = []
high_quality_weibo2 = []
high_quality_id = []
high_quality_time = []
high_quality_tag = []
for each in itw1:
high_quality_id.append(each[0])
high_quality_time.append(str(each[1]))
high_quality_weibo.append(each[2])
high_quality_weibo2.append(each[3])
high_quality_tag.append(each[4])
quick_write_list_to_text(high_quality_id,
write_directory + '/weibo_id.txt')
quick_write_list_to_text(high_quality_time,
write_directory + '/weibo_time.txt')
quick_write_list_to_text(high_quality_weibo,
write_directory + '/weibo_content.txt')
quick_write_list_to_text(high_quality_weibo2,
write_directory + '/weibo_content2.txt')
quick_write_list_to_text(high_quality_tag,
write_directory + '/weibo_class_tag.txt')
quick_write_list_to_text(all_weibo_word,
write_directory + '/weibo_word.txt')
