def fuzzyQuery(query, index):
"""find the word in index which is similar to query"""
# sim keywords sequences set
simKeywordsSeq = set()
# input matrix
mat = [Unigram.bitArrayGen(query)]
for keyword in index:
mat.append(Unigram.bitArrayGen(keyword))
# transpose the matrix
mat = np.array(mat).T
hashBucket = min_hash.minHash(mat, 20, 5)
# the first col denotes query
for i in hashBucket:
if 0 in hashBucket[i]:
simKeywordsSeq.update(hashBucket[i])
# remove the query seq
simKeywordsSeq.remove(0)
return [index[i - 1] for i in simKeywordsSeq]
import sys
import question as q
import unigram as u
import Unigram as U
import entropy as e
udic = U.getDictNew("trainB")
uentropy=e.entropy(udic)
print uentropy
result=[]
text = [line.strip() for line in open("../data/trainB.txt")]
question=q.getDic()
for i in range(1, 505):
print "question "+str(i)
#data=[]
dic1={}
dic2={}
count=0
c1=0
c2=0
if i in question and i!=146:
for line in text:
word = line.split(" ")
sent=word[:-1]
w=word[-1]
#for j in range(len(word)):
count+=1
#data.append((sent, word[i]))
if q.askQuestion(i, question[i], sent):
c1+=1
if w in dic1:
import sys
import question as q
import unigram as u
import Unigram as U
import entropy as e
udic = U.getDictNew("trainA")
uentropy=e.entropy(udic)
print uentropy
result=[]
text = [line.strip() for line in open("../data/trainA.txt")]
question=q.getDic()
print "*********************"
print uentropy
print "*********************"
for i in range(1, 505):
print "question "+str(i)
#data=[]
dic1={}
dic2={}
count=0
c1=0
c2=0
if i in question:
for line in text:
word = line.split(" ")
sent=word[:-1]
w=word[-1]
#for j in range(len(word)):
count+=1
#data.append((sent, word[i]))
