#!/usr/bin/env python
# coding: utf-8
#
# Author: Peinan ZHANG
# Created at: 2014-10-24
import sys
from q010_sort_second_wFreq import sortSecWFreq
from collections import defaultdict
def calcBigramFreq(lines, unigram_freq):
bigram_freq = {}
for line in lines:
count, c_word, n_word = line.strip().decode('utf-8').split('\t')
bigram_freq['%s\t%s' % (c_word, n_word)] = \
float(count) / unigram_freq[c_word]
return bigram_freq
if __name__ == '__main__':
unigram_freq = sortSecWFreq('data/medline.txt.sent.tok')
bigram_freq = calcBigramFreq(sys.stdin.readlines(), unigram_freq)
for k, v in sorted(bigram_freq.items(), key=lambda x: x[1], reverse=True):
sys.stdout.write('%f\t%s\n' % (v, k.encode('utf-8')))
#!/usr/bin/env python
# coding: utf-8
#
# Author: Peinan ZHANG
# Created at: 2014-10-23
import sys
def topN(freqDict, N=100, c=True):
from q010_sort_second_wFreq import sortSecWFreq
count = 0
return_list = []
for k, v in sorted(freqDict.items(), key=lambda x: x[1], reverse=True):
if count < N:
if c == True:
return_list.append('%3d %s\n' % (v, k.encode('utf-8')))
if c == False:
return_list.append('%s\n' % k.encode('utf-8'))
count += 1
else:
break
return return_list
if __name__ == '__main__':
for item in topN(sortSecWFreq(sys.argv[1])):
sys.stdout.write(item)
#!/usr/bin/env python
# coding: utf-8
#
# Author: Peinan ZHANG
# Created at: 2014-10-23
import sys, os
from q027_top100 import topN
from q010_sort_second_wFreq import sortSecWFreq
def mkBigramFile(filepath):
with open('bigram.temp', 'w') as bigramFile:
for line in open(filepath):
line = line.strip().decode('utf-8')
if len(line) <= 2:
bigramFile.write(line.encode('utf-8') + '\n')
continue
for i in range(len(line) - 1):
bigramFile.write('%s%s\n' % \
(line[i].encode('utf-8'), line[i + 1].encode('utf-8')))
if __name__ == '__main__':
mkBigramFile(sys.argv[1])
for item in topN(sortSecWFreq('bigram.temp')):
sys.stdout.write(item)
os.remove('bigram.temp')