-
Notifications
You must be signed in to change notification settings - Fork 1
/
bayesianClassifier.py
333 lines (287 loc) · 9.08 KB
/
bayesianClassifier.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
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
import re
import commands
import pickle
import math
import nltk
import text
import time
import operator
def collocations(set_of_words):
global contextWords
correct = {}
correctset = {}
line = ""
for eve in set_of_words[0]:
line = line + " " + eve
word_tok = nltk.word_tokenize(line)
pos_tags = nltk.pos_tag(word_tok)
for words in set_of_words:
#print(words)
for i in range(0, len(words)):
words[i] = words[i].lower()
for i in range(0, len(words)):
if words[i] in cWords:
start = max(0,i-3)
end = min(i+3, len(words)-1)
left_seq = ''
for j in range(start,i):
left_seq = left_seq + pos_tags[j][1] + ','
right_seq = ''
for j in range(i+1,end+1):
right_seq = right_seq + pos_tags[j][1] + ','
prob = {}
confuse = cWords[words[i]]
for w in confuse:
prob[w] = priorConf[w]
if left_seq in collocs_left[w]:
val_l = float(collocs_left[w][left_seq]+1)/(confCounts[w] + len(collocs_left[w]))
else:
val_l = 1
if right_seq in collocs_right[w]:
val_r = float(collocs_right[w][right_seq]+1)/(confCounts[w] + len(collocs_right[w]))
else:
val_r = 1
prob[w]*=(val_r*val_l)
maxval = 0
idx = ''
#for p in prob:
# print(p + ' '),
# print(prob[p])
# for k in prob:
# if prob[k] > maxval:
# maxval = prob[k]
# idx = k
# print(k)
normalize = 0
for k in prob:
normalize += prob[k]
maxvalue = 0
maxword = ""
for k in prob:
if maxvalue < prob[k]:
maxvalue = prob[k]
maxword = k
prob[k] /= normalize
correctset[maxword] = maxvalue
for eve in prob:
if eve in correct:
correct[eve] += prob[eve]
else:
correct[eve] = prob[eve]
minvalue = 1000
minword = ""
actual = {}
if len(correct) > 0:
for eve in correctset:
if minvalue > correct[eve]:
minvalue = correct[eve]
minword = eve
actuallist = cWords[minword]
for eve in actuallist:
actual[eve] = correct[eve]
else:
actual = correct
#print actual
return actual
def context(set_of_words):
global contextWords
correct = {}
correctset = {}
for words in set_of_words:
for i in range(0, len(words)):
words[i] = words[i].lower()
for i in range(0, len(words)):
if words[i] in cWords:
start = max(0,i-3)
end = min(i+3, len(words)-1)
context = set()
for j in range(start, end+1):
context.add(words[j])
prob = {}
confuse = cWords[words[i]]
for w in confuse:
prob[w] = priorConf[w]
for c in context:
val = 1
if c in contextWords[w]:
val = (float(contextWords[w][c])+1)/(1.0*(confCounts[w] + len(contextWords[w])))
#val = (float(contextWords[w][c]))
else:
val = 1
prob[w] *= val
maxval = 0
idx = ''
#for p in prob:
# print(p + ' '),
# print(prob[p])
#for k in prob:
# if prob[k] > maxval:
# maxval = prob[k]
# idx = k
# print(k)
normalize = 0
for k in prob:
normalize += prob[k]
maxvalue = 0
maxword = ""
for k in prob:
if maxvalue < prob[k]:
maxvalue = prob[k]
maxword = k
prob[k] /= normalize
correctset[maxword] = maxvalue
for eve in prob:
if eve in correct:
correct[eve] += prob[eve]
else:
correct[eve] = prob[eve]
minvalue = 1000
minword = ""
actual = {}
if len(correct) > 0:
for eve in correctset:
if minvalue > correct[eve]:
minvalue = correct[eve]
minword = eve
actuallist = cWords[minword]
for eve in actuallist:
actual[eve] = correct[eve]
else:
actual = correct
#print actual
return actual
contextWords = pickle.load(open('data/brown/ContextWords.dict', 'rb'))
collocs_left = pickle.load(open('data/brown/Collocs_left_pruned.dict', 'rb'))
collocs_right = pickle.load(open('data/brown/Collocs_right_pruned.dict', 'rb'))
confusionSets = pickle.load(open('data/brown/ConfusionSets.dict', 'rb'))
fp = open("data/all-words-cleaned.txt", 'r')
data = fp.read()
fp.close()
splitdata = data.split('\n')
numOfWords = len(splitdata) - 1
dicti = []
for i in range(0, numOfWords):
temp = splitdata[i].lower()
dicti.append(temp)
cWords = {}
confCounts = {}
priorConf = {}
for cSet in confusionSets:
sum_set = 0
for cWord in confusionSets[cSet]:
sum_set += confusionSets[cSet][cWord]
cWords[cWord] = confusionSets[cSet]
for cWord in confusionSets[cSet]:
priorConf[cWord] = float(confusionSets[cSet][cWord])/float(sum_set)
confCounts[cWord] = float(confusionSets[cSet][cWord])
#for w in priorConf:
# print(w+ ' '),
# print(priorConf[w])
#print(confCounts)
#print(priorConf)
def bayesianCheck(line):
#print line + '\t',
usedSpell = False
line = line.lower()
line = line.replace('\n', '')
line = re.sub('[^0-9a-zA-Z ]+', '', line)
words_ = line.split(' ')
set_of_words = []
minprob = 1
minidx = 0
poss_list = {}
maxprob = 0
for i in range(0, len(words_)):
temp = ''
maxval = 0
poss = {}
if words_[i] not in dicti:
#print words_[i]
usedSpell = True
poss = text.correctWord(words_[i])
else:
poss[words_[i]] = 1
continue
for p in poss:
if poss[p] > maxval:
maxval = poss[p]
temp = p
if len(temp) == 0:
temp = words_[i]
#poss[temp] = 1
else:
words_[i] = temp
if maxprob < poss[temp]:
maxidx = i
maxprob = poss[temp]
poss_list = poss
kter = 0
if len(poss_list) == 0:
set_of_words.append([])
for wter in words_:
set_of_words[kter].append(wter)
else:
for w in poss_list:
if (poss_list[w] > 0.2):
set_of_words.append([])
for wter in words_:
set_of_words[kter].append(wter)
set_of_words[kter][minidx] = w
kter += 1
con = context(set_of_words)
col = collocations(set_of_words)
#print con
#print col
#print "One Done #########################################"
total = {}
final = 0
for eve in con:
total[eve] = con[eve]
final += con[eve]
for eve in col:
if eve not in total:
total[eve] = col[eve]
else:
total[eve] += col[eve]
final += col[eve]
for eve in total:
total[eve] = total[eve]/final
value = 0
coreve = ""
if usedSpell:
for eve in poss_list:
if value < poss_list[eve]:
value = poss_list[eve]
coreve = eve
#if coreve in total:
# temp = cWords[coreve]
# for x in temp:
#print x + " " + str(total[x])
sorted_p = sorted(poss_list.items(), key=operator.itemgetter(1), reverse = True)
#counter = 0
#output = {}
#for fin in sorted_p:
# if counter == 3:
# break
#print fin[0] + "\t" + str(fin[1]),
# counter += 1
# output[fin[0]] = fin[1]
return sorted_p
#print
else:
sorted_t = sorted(total.items(), key=operator.itemgetter(1), reverse = True)
#counter = 0
return sorted_t
#output = {}
#for fin in sorted_t:
# if counter == 3:
# break
# output[fin[0]] = fin[1]
#print fin[0] + "\t" + str(fin[1]),
# counter += 1
#print
#return output
#while(True):
# line = raw_input('Enter line:')
# output = bayesianCheck(line)
# print(output)