forked from eric-sai/Data-Mining-on-Twitter
/
proj2.py
210 lines (199 loc) · 8 KB
/
proj2.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
#main running file. All the procesure are running here
import math
from math import sqrt
import nltk
from index import make_invidx
from index import make_invidx_by_OkapiBM25
from index import make_index_AIG
from query import get_query
from query import calculate_query
from query import cosine_distance
from query import expand_query
from query import pivoted_length
from query import score
from query import sum_keyword
from storage import saveIndex
from storage import loadIndex
from evaluation import readFile
from evaluation import precision
from evaluation import recall
from evaluation import printEva
from evaluation import MAP
from evaluation import Avp
from index import make_invidx_by_OkapiBM251
from index import make_invidx_by_OkapiBM25s
from index import read_file
from queryExapnsion import TSV
from query import sum_score
from userFeedback import userFeedback
from userFeedback import coWeight
from userFeedback import selectTopTerm
from evaluation import readFile
from evaluation import precision
from evaluation import MRR
import time
#file="C://TweetStudy//12//12//00_00.dat"
queryAddress="C://TweetStudy//12//12//query.txt"
documentAddress="C://TweetStudy//12//12/2"
extend={"*.dat"}
#extend={"00*.dat","01*dat","02*dat","03*.dat","04*dat","05*.dat","06*.dat","07*dat","08*dat","09*.dat"}
#extend={"10*.dat","11*dat","12*dat","13*.dat","14*dat","15*.dat","16*.dat","17*dat","18*dat","19*.dat"}
#extend={"20*.dat","21*.dat","22*.dat","23*.dat"}
index = "C://TweetStudy//12//12//00_query.txt"
resultAddress="C://TweetStudy//12//12//relevantList.txt"
def get_k(queryAddress,documentAddress,e):
document1, length_by_docid1 = make_invidx(documentAddress, e)
query1=calculate_query(get_query(queryAddress))
simi1= cosine_distance(document1,query1,100)
result=readFile(resultAddress)
for num in simi1:
for weight1,docid in simi1[num]:
if docid in result:
print(docid)
def get_topK(queryAddress,documentAddress,extend):
for e in extend:
document1, length_by_docid1 = make_invidx(documentAddress, e)
document2 = make_invidx_by_OkapiBM25(documentAddress, e)
#print(document)
#print(length_by_docid)
#saveIndex(index, (document, length_by_docid))
#document, length_by_docid = loadIndex(index)
#get query
query1=calculate_query(get_query(queryAddress))
for num in query1:
document3= make_index_AIG(documentAddress,e,query1[num])
simi3=sum_score(document3,10)
simi2=sum_keyword(document2,query1[num],10)
#print(simi)
#document= make_index_AIG(documentAddress,e,query1[num])
simi1= cosine_distance(document1,query1,10)
# simi=sum_keyword(document, query1[num],10)
#print(query1)
#cosine similarity
#simi=score(document,query1,10)
#simi= cosine_distance(document,query1,10)
#print(simi)
for num in simi1:
for weight1,docid in simi1[num]:
#print(docid)
for weight2,docid2 in simi2:
if docid==docid2:
print(docid2," : ",weight2)
for weight3,docid3 in simi3:
if docid==docid3:
print(docid3,":",weight3)
for weight2,docid2 in simi2:
for weight3,docid3 in simi3:
if docid2==docid3:
print(docid3,":",weight3)
# print(num, docid, ":",weight)
def queryExpand1(document,extend,queryAddress):
invx ,tf_by_docid,df,N= make_invidx_by_OkapiBM251(document, extend)
query1=calculate_query(get_query(queryAddress))
simi4={}
for num in query1:
simi=sum_keyword(invx,query1[num],20)
termSelected=TSV(tf_by_docid,df,N,simi)
#print(termSelected)
invx1=make_invidx_by_OkapiBM25s(document, extend,termSelected)
a= sorted([(termSelected[term], term) for term in termSelected], reverse=True)[:10]
print(a)
for weight,term in a:
query1[num][term]=weight
simi4=sum_keyword(invx1,query1[num],1000)
#print(simi4)
return simi4
def userLogExpan(extend,documentAddress,queryAddress):
ses=userFeedback("C://TweetStudy//12//12//userFeedback3.txt")
# print(ses)
ret={}
for e in extend:
document1, length_by_docid1 = make_invidx(documentAddress, e)
tf,atid=read_file(documentAddress,e)
# query1=calculate_query(get_query(queryAddress))
dic=get_query(queryAddress)
for num in dic:
re=coWeight(ses,nltk.word_tokenize(dic[num]),document1,tf,atid)
ret=selectTopTerm(re,10)
return ret
def evaluation(extend, documentAddress, queryAddress,resultAddress):
result=readFile(resultAddress)
#print(result)
#tf-idf
for e in extend:
# document1, length_by_docid1 = make_invidx(documentAddress, e)
query1=calculate_query(get_query(queryAddress))
# simi1= cosine_distance(document1,query1,1000)
retrieval =0.0
relevance=0.0
document2 = make_invidx_by_OkapiBM25(documentAddress, e)
for num in query1:
# print("tf-idf: ")
# print(MRR(result, simi1[num]))
#print(recall(result,simi1[num]))
# document3= make_index_AIG(documentAddress,e,query1[num])
# simi3=sum_score(document3,1000)
# print("AIG: ")
# print(MRR(result, simi3))
#print(recall(result,simi3))
simi2=sum_keyword(document2,query1[num],1000)
print("BM25: ")
print(MRR(result, simi2))
#print(recall(result,simi2))
# for num in simi1:
# for weight1,docid in simi1[num]:
# #print(docid)
# for weight2,docid2 in simi2:
# if docid==docid2:
# retrieval = retrieval + 1
# if docid in result:
# relevance=relevance + 1
# print(relevance/retrieval)
def queryExpanEva1(extend,documentAddress,queryAddress,resultAddress):
for e in extend:
sim4=queryExpand1(documentAddress,e,queryAddress)
result=readFile(resultAddress)
print("precision:")
print(MRR(result, sim4))
# print("recall:")
# print(recall(result,sim4))
def queryExpanEva(extend,documentAddress,queryAddress,resultAddress):
expandedTerm=userLogExpan(extend,documentAddress,queryAddress)
# expandedTerm=queryExpand1(document,extend,queryAddress)
print(expandedTerm)
query1=calculate_query(get_query(queryAddress))
result=readFile(resultAddress)
for num in query1:
for term in expandedTerm:
query1[num][term]=query1[num].get(term,0) + 1
#print(query1)
for e in extend:
# document1, length_by_docid1 = make_invidx(documentAddress, e)
# simi1= cosine_distance(document1,query1,1000)
document2 = make_invidx_by_OkapiBM25(documentAddress, e)
# simi1= cosine_distance(document1,query1,50)
# print(simi1)
for num in query1:
# print("tf-idf: ")
# print(MRR(result, simi1[num]))
# print(recall(result,simi1[num]))
# document3= make_index_AIG(documentAddress,e,query1[num])
# simi3=sum_score(document3,1000)
simi2=sum_keyword(document2,query1[num],1000)
# print("AIG: ")
# print(MRR(result, simi3))
# print(recall(result,simi3))
print("BM25: ")
print(MRR(result, simi2))
# print(recall(result,simi2))
#get_topK(queryAddress,documentAddress,extend)
#for e in extend:
# queryExpand(documentAddress,e,queryAddress)
#get_k(queryAddress,documentAddress,extend)
evaluation(extend,documentAddress,queryAddress,resultAddress)
queryExpanEva(extend,documentAddress,queryAddress,resultAddress)
queryExpanEva1(extend,documentAddress,queryAddress,resultAddress)
documentAddress="C://TweetStudy//12//12//1"
evaluation(extend,documentAddress,queryAddress,resultAddress)
queryExpanEva(extend,documentAddress,queryAddress,resultAddress)
queryExpanEva1(extend,documentAddress,queryAddress,resultAddress)