-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdemo.py
1098 lines (982 loc) · 53.8 KB
/
demo.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
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
"""
网站Demo API模块
"""
# -*- coding: UTF-8 -*-
# import os
import sys
import numpy as np
import imp
import json
import random
# import matplotlib.pyplot as plt
from flask import Flask, request, redirect, url_for
from flask import render_template
from flask_bootstrap import Bootstrap
from flask_wtf import FlaskForm
from flask_wtf.file import FileField
from flask_uploads import UploadSet, configure_uploads, TEXT, patch_request_class, UploadNotAllowed
from wtforms import SubmitField, TextAreaField
# from collections import defaultdict
# from werkzeug.utils import secure_filename
from Fine_grained.Src.Fine_grained.CONFIG import CONF
from psutil import cpu_count
# from fine_grained import init, analysis_comment
from summary import gen_summary
# 不显示server输出
import logging
from Fine_grained.Src.Fine_grained.FINE_GRAINED import init_knowledge_base, analysis_comment
from global_var import gl
app = Flask(__name__)
bootstrap = Bootstrap(app)
app.config['SECRET_KEY'] = 'AIMindreader'
app.config['UPLOADED_TEXTS_DEST'] = './uploads'
UPLOAD_FOLDER = './uploads'
ALLOWED_EXTENSIONS = set(['txt', 'statelist'])
thread_num=cpu_count(logical=False)
texts = UploadSet('texts', TEXT)
configure_uploads(app, texts)
patch_request_class(app) # 文件大小限制，默认为16MB
def init():
# import gensim
gl.set_value('PRODUCT', '汽车')
gl.set_value('STDOUT', False)
gl.set_value('PROGRESS', 0) # 后端分析进度百分比
gl.set_value('STATE', 'free')
# gl.set_value('WORD2VEC_MODEL', gensim.models.Word2Vec.load(CONF.WORD2VEC_PATH))
init_data = init_knowledge_base()
gl.set_value('PROFILE_TREE', None)
gl.set_value('ENT_ATTR_POLAR', None)
gl.set_value('ENT_ATTR_TEXT', None)
gl.set_value('ENT_POLAR', None)
gl.set_value('ENT_POLAR_INCLUDE_CHILDREN', None)
gl.set_value('INIT_DATA', init_data)
gl.set_value('SORTED_UNIQUE_WORDS', set())
gl.set_value('SORTED_UNIQUE_WORDS_ENTITIES', set())
gl.set_value('SORTED_UNIQUE_WORDS_ATTRIBUTES', set())
gl.set_value('SORTED_UNIQUE_WORDS_VA', set())
imp.reload(sys)
sys.path.append("..")
log = logging.getLogger('werkzeug')
log.setLevel(logging.ERROR)
# from utils.misc_utils import get_args_info
@app.route('/changePolar', methods=['GET', 'POST'])
def changePolar():
attribute=request.args.get('attribute')
description=request.args.get('description')
polar=request.args.get('polar')
try:
supplement_file=open(CONF.SUPPLEMENT_ATTRIBUTE_DESCRIPTION_PATH,'a+',encoding='utf8')
supplement_file.write("\n%s\t%s\t%s"%(attribute,polar,description))
supplement_file.close()
response='感谢您的贡献，让我们的知识库变得更加精准！'
except Exception as e:
response='修正过程出现异常，请点击右上角 导航->联系我们 反馈此问题'
raise(e)
finally:
return response
@app.route('/send_mail', methods=['GET', 'POST'])
def send_mail():
import smtplib
from email.mime.text import MIMEText
from email.header import Header
from email.utils import formataddr
file = open('CONFIG_MAIL.dat', encoding='utf-8')
mail_user = file.readline().replace('\n', '').replace('\r', '')
mail_pass = file.readline().replace('\n', '').replace('\r', '')
file.close()
mail_host = "smtp." + mail_user.split('@')[1] # 设置服务器
sender = request.args.get('sender')
name = request.args.get('name')
# sender='习近平@126.com'
subject = request.args.get('subject')
# subject='同志们好'
text = request.args.get('text')
# text='我就无聊了群发个邮件'
message = MIMEText(text, 'plain', 'utf-8')
message['From'] = formataddr((Header(name, "utf-8").encode(), sender))
message['To'] = formataddr((Header("网站系统反馈", 'utf-8').encode(), mail_user))
message['Subject'] = Header(subject, 'utf-8')
try:
smtpObj = smtplib.SMTP()
# smtpObj.set_debuglevel(1)
# smtpObj.ehlo(mail_host)
smtpObj.connect(mail_host)
smtpObj.login(mail_user.split('@')[0], mail_pass)
smtpObj.sendmail(mail_user, [mail_user], message.as_string())
# smtpObj.quit()
print("邮件发送成功")
return '1'
except smtplib.SMTPException:
print("邮件发送异常")
return '-1'
def word_cloud_generate(upload_file_name, in_img_path=u'static/WebTemplate/images/car.jpg',
font_path=r'static/WebTemplate/fonts/msyh.ttc'):
import jieba
from PIL import Image
from wordcloud import WordCloud, STOPWORDS
in_text_path = u'uploads/' + upload_file_name
out_img_path = u'static/result_wordCloud/' + upload_file_name + '.png'
f = open(in_text_path, 'r', encoding='utf8').read()
coloring = np.array(Image.open(in_img_path))
stopwords = set(STOPWORDS)
stopwords.update(
["没有", "感觉", "可以", "现在", "公里", "就是", "比较", "应该", "大家", "左右", "居然", "不是", "还是", "不过", "空间", "一次", "什么", "不到",
"那个", "东西", "这个", "一下", "这么", "怎么", "一直", "时候", "时间", "次", "之后", "店", "眼", "问题", "毛病", "地方", "原因", "感受", "现象",
"事", "上", "下", "范围", "朋友", "缺点", "优点", "车友", "车主", "第一", "一点", "水平", "程度"])
wordlist = jieba.cut(f, cut_all=True)
wordlist = " ".join(wordlist)
wordcloud = WordCloud(background_color="white", font_path=font_path, mask=coloring, stopwords=stopwords, width=1000,
height=860, margin=2).generate(wordlist)
# image_colors = ImageColorGenerator(coloring)
# plt.imshow(wordcloud, interpolation="bilinear")
# plt.axis("off")
# plt.show()
wordcloud.to_file(out_img_path)
class SimpleGroupedColorFunc(object):
"""Create a color function object which assigns EXACT colors
to certain words based on the color to words mapping
Parameters
----------
color_to_words : dict(str -> list(str))
A dictionary that maps a color to the list of words.
default_color : str
Color that will be assigned to a word that's not a member
of any value from color_to_words.
"""
def __init__(self, color_to_words, default_color):
self.word_to_color = {word: color
for (color, words) in color_to_words.items()
for word in words}
self.default_color = default_color
def __call__(self, word, **kwargs):
return self.word_to_color.get(word, self.default_color)
def word_cloud_generate(in_img_path=u'static/WebTemplate/images/car.jpg',
font_path=r'static/WebTemplate/fonts/msyh.ttc'):
from PIL import Image
from wordcloud import WordCloud
ent_polar_include_children = gl.get_value('ENT_POLAR_INCLUDE_CHILDREN')
word_freq_polars = gl.get_value('WORD_FREQ')
word_freq = dict()
color2words = dict()
for word, freq_polar in word_freq_polars.items():
freq = sum(freq_polar)
if freq == 0:
continue
word_freq[word] = freq
color = polar2color(freq_polar)
words = color2words.setdefault(color, [])
words.append(word)
grouped_color_func = SimpleGroupedColorFunc(color2words, default_color='grey')
out_img_path = u'static/result_wordCloud/' + gl.get_value('UPLOAD_FILE_PATH') + '.png'
coloring = np.array(Image.open(in_img_path))
wordcloud = WordCloud(background_color="white", font_path=font_path, mask=coloring, width=1000,
height=860, margin=2).generate_from_frequencies(word_freq)
wordcloud.recolor(color_func=grouped_color_func)
# image_colors = ImageColorGenerator(coloring)
# plt.imshow(wordcloud, interpolation="bilinear")
# plt.axis("off")
# plt.show()
wordcloud.to_file(out_img_path)
def polar2color(polar):
pos = polar[0]
neg = polar[1]
if neg + pos == 0:
return '#ffff00'
r = neg / (neg + pos)
g = pos / (neg + pos)
return '#' + hex(int(r * 255))[2:].rjust(2, '0') + hex(int(g * 255))[2:].rjust(2, '0') + '00'
def knowledge_base_init():
product = gl.get_value('PRODUCT', '汽车')
entity_pair, entity_level = read_aspect_file('./KnowledgeBase/' + product + '/whole-part.txt')
entity_attr, _ = read_aspect_file('./KnowledgeBase/' + product + '/entity-attribute.txt', entity_level)
entity_synonym, _ = read_aspect_file('./KnowledgeBase/' + product + '/entity-synonym.txt', entity_level)
attr_synonym, _ = read_aspect_file('./KnowledgeBase/' + product + '/attribute-synonym.txt')
opinion_pair = read_opinion_file('./KnowledgeBase/' + product + '/attribute-description.txt')
subset = read_subset_file('./KnowledgeBase/' + product + '/subset.txt')
entity_tree = {'name': product, 'child': [], 'id': 1, 'type': 'entity'}
kb_build_entity(entity_tree, entity_pair, 1)
gl.set_value('ENTITY_PAIR', entity_pair)
gl.set_value('ENTITY_LEVEL', entity_level)
gl.set_value('ENTITY_ATTRIBUTE', entity_attr)
gl.set_value('ENTITY_SYNONYM', entity_synonym)
gl.set_value('ATTRIBUTE_SYNONYM', attr_synonym)
gl.set_value('OPINION_PAIR', opinion_pair)
gl.set_value('ENTITY_TREE', entity_tree)
gl.set_value('SUBSET', subset)
gl.set_value('SORTED_UNIQUE_WORDS', set())
gl.set_value('SORTED_UNIQUE_WORDS_ENTITIES', set())
gl.set_value('SORTED_UNIQUE_WORDS_ATTRIBUTES', set())
gl.set_value('SORTED_UNIQUE_WORDS_VA', set())
gl.set_value('UNLABELED_TEXT', [])
def save_unlabeled_text(unlabeled_text, file_path):
if len(unlabeled_text) == 0:
return
f = open(file_path, 'w', encoding='utf8')
f.write('\n'.join(unlabeled_text))
f.close()
def flush_unlabeled_text():
import os, datetime
unlabeled_text = gl.get_value('UNLABELED_TEXT', [])
save_unlabeled_text(unlabeled_text, CONF.UNLABELED_TEXT_PATH)
gl.set_value('UNLABELED_TEXT',[])
product = gl.get_value('PRODUCT', '汽车')
CONF.UNLABELED_TEXT_PATH = os.path.abspath(
'./UnlabeledText' + '/' + product + '/' + datetime.datetime.now().strftime('%Y%m%d_%H%M%S') + '.txt')
def knowledge_graph():
product = gl.get_value('PRODUCT', '汽车')
subset = gl.get_value('SUBSET', None)
if subset is None:
knowledge_base_init()
subset = gl.get_value('SUBSET')
file = open('static/kb_json/' + product + '/subset.js', 'wb')
file.write(
('subset=\'' + json.dumps(subset) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
subset_nodes = subset['nodes']
links = {'entity_entity': [], 'entity_attribute': [], 'entity_synonym': [], 'attribute_synonym': [],
'attribute_opinion': []}
nodes = dict()
attribute_level = 5
opinion_level = 6
# entity-entity
entity_pair = gl.get_value('ENTITY_PAIR', None)
if entity_pair is None:
knowledge_base_init()
entity_pair = gl.get_value('ENTITY_PAIR')
for parent, child, parent_level in entity_pair:
if parent not in subset_nodes or child not in subset_nodes:
continue
links['entity_entity'].append(
{'source': parent, 'target': child, 'type': 'is part of', 'source_level': parent_level,
'target_level': parent_level + 1})
if nodes.get(parent, None) is None:
nodes[parent] = {'name': parent, 'type': 'entity', 'level': parent_level}
if nodes.get(child, None) is None:
nodes[child] = {'name': child, 'type': 'entity', 'level': parent_level + 1}
# entity-attribute
entity_attribute = gl.get_value('ENTITY_ATTRIBUTE')
for entity, attribute, _ in entity_attribute:
if entity not in subset_nodes or attribute not in subset_nodes:
continue
links['entity_attribute'].append({'source': entity, 'target': attribute, 'type': 'is an attribute of',
'source_level': nodes[entity]['level'], 'target_level': attribute_level})
if nodes.get(attribute, None) is None:
nodes[attribute] = {'name': attribute, 'type': 'attribute', 'level': attribute_level}
# entity-synonym
entity_synonym = gl.get_value('ENTITY_SYNONYM')
for entity, synonym, _ in entity_synonym:
if entity in subset_nodes and nodes.get(synonym, None) is None:
entity_level = nodes[entity]['level']
links['entity_synonym'].append(
{'source': entity, 'target': synonym, 'type': 'is the same as', 'source_level': entity_level,
'target_level': entity_level + 1})
nodes[synonym] = {'name': synonym, 'type': 'entity_synonym', 'level': entity_level + 1}
# attribute-synonym
attribute_synonym = gl.get_value('ATTRIBUTE_SYNONYM')
for attribute, synonym, _ in attribute_synonym:
if attribute in subset_nodes and nodes.get(synonym, None) is None:
links['attribute_synonym'].append(
{'source': attribute, 'target': synonym, 'type': 'is the same as', 'source_level': attribute_level,
'target_level': attribute_level + 1})
nodes[synonym] = {'name': synonym, 'type': 'attribute_synonym', 'level': attribute_level + 1}
# attribute-opinion
opinion_pair = gl.get_value('OPINION_PAIR')
for attribute, opinion, polar in opinion_pair:
if attribute not in subset_nodes or opinion not in subset_nodes:
continue
type = None
if polar == 1:
type = 'positive description'
elif polar == -1:
type = 'negative description'
else:
type = 'neutral description'
links['attribute_opinion'].append(
{'source': attribute, 'target': opinion, 'type': 'is a ' + type + ' of', 'source_level': attribute_level,
'target_level': opinion_level})
if nodes.get(opinion, None) is None:
nodes[opinion] = {'name': opinion, 'type': type, 'level': opinion_level}
gl.set_value('KNOWLEDGE_GRAPH_LINKS', links)
gl.set_value('KNOWLEDGE_GRAPH_NODES', nodes)
# if not os.path.exists('static/kb_json/'+product+'/part_of_knowledgebase.js'):
file = open('static/kb_json/' + product + '/part_of_knowledgebase.js', 'wb')
file.write(
('links_kb_subset=\'' + json.dumps(links) + '\';\nnodes_kb_subset=\'' + json.dumps(nodes) + '\';').replace(
'\\"', '\\\\"').encode('utf-8'))
file.close()
def kb_build_entity(node, entity_pair, id):
for parent, child, _ in entity_pair:
if parent == node['name']:
id = id + 1
node['child'].append({'name': child, 'child': [], 'id': id, 'type': 'entity'})
for child in node['child']:
id = kb_build_entity(child, entity_pair, id)
return id
def kb_build_level2(lv1_lv2, root_name, lv1_type, lv2_type):
root = {'name': root_name, 'child': [], 'id': 1, 'type': lv1_type}
id = 1
for lv1, lv2, _ in lv1_lv2:
if lv1 != root_name:
continue
children = root['child']
if lv2 not in children:
id = id + 1
children.append({'name': lv2, 'child': [], 'id': id, 'type': lv2_type})
return root
def kb_build_attr_opnion(attr_opinion, attr_name):
root = {'name': attr_name, 'child': [], 'id': 1, 'type': 'attribute'}
pos_node = {'name': '正向描述', 'child': [], 'id': 2, 'type': 'describe'}
neu_node = {'name': '中性描述', 'child': [], 'id': 3, 'type': 'describe'}
neg_node = {'name': '负向描述', 'child': [], 'id': 4, 'type': 'describe'}
id = 4
for attr, opinion, polar in attr_opinion:
if attr != attr_name:
continue
id = id + 1
if polar == 1:
pos_node['child'].append({'name': opinion, 'child': [], 'id': id, 'type': 'opinion', 'polar': 1})
elif polar == -1:
neg_node['child'].append({'name': opinion, 'child': [], 'id': id, 'type': 'opinion', 'polar': -1})
else:
neu_node['child'].append({'name': opinion, 'child': [], 'id': id, 'type': 'opinion', 'polar': 0})
root['child'].append(pos_node)
root['child'].append(neu_node)
root['child'].append(neg_node)
return root
@app.route('/index', methods=['GET', 'POST'])
def home():
return render_template('home.html')
@app.route('/contact', methods=['GET', 'POST'])
def contact():
return render_template('contact.html')
@app.route('/introduction', methods=['GET', 'POST'])
def introduction():
product = gl.get_value('PRODUCT', '汽车')
# if not os.path.exists('static/kb_json/'+product+'/part_of_knowledgebase.js'):
knowledge_graph()
return render_template('introduction.html')
@app.route('/knowledge_base', methods=['GET', 'POST'])
def kb_graph():
ent = request.args.get('entity')
attr = request.args.get('attribute')
if ent is None:
ent = '0'
if attr is None:
attr = '0'
entity_tree = gl.get_value('ENTITY_TREE', None)
if entity_tree is None:
knowledge_base_init()
entity_tree = gl.get_value('ENTITY_TREE')
product = gl.get_value('PRODUCT', '汽车')
# if not os.path.exists('static/kb_json/'+product+'/whole_part.js'):
file = open('static/kb_json/' + product + '/whole_part.js', 'wb')
file.write(('whole_part=\'' + json.dumps(entity_tree) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
if ent != '0':
# if not os.path.exists('static/kb_json/'+product+'/' + ent + '.js'):
ent_attr = kb_build_level2(gl.get_value('ENTITY_ATTRIBUTE'), ent, 'entity', 'attribute')
ent_synonym = kb_build_level2(gl.get_value('ENTITY_SYNONYM'), ent, 'entity', 'entity_synonym')
file = open('static/kb_json/' + product + '/' + ent + '.js', 'wb')
file.write(('ent_attr=\'' + json.dumps(ent_attr) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.write(('ent_synonym=\'' + json.dumps(ent_synonym) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
else:
ent = None
if attr != '0':
# if not os.path.exists('static/kb_json/'+product+'/' + attr + '.js'):
attr_opinion = kb_build_attr_opnion(gl.get_value('OPINION_PAIR'), attr)
attr_synonym = kb_build_level2(gl.get_value('ATTRIBUTE_SYNONYM'), attr, 'attribute', 'attribute_synonym')
file = open('static/kb_json/' + product + '/' + attr + '.js', 'wb')
file.write(('attr_opinion=\'' + json.dumps(attr_opinion) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.write(('attr_synonym=\'' + json.dumps(attr_synonym) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
else:
attr = None
return render_template('knowledge_base.html', ent=ent, attr=attr, product=product)
@app.route('/getProduct', methods=['GET', 'POST'])
def get_product():
return gl.get_value('PRODUCT', '汽车')
@app.route('/changeProduct', methods=['GET', 'POST'])
def change_product():
product = request.args.get('product')
gl.set_value('PRODUCT', product)
flush_unlabeled_text()
# init_data = init(use_nn=use_nn)
CONF.__init__(gl.get_value('PRODUCT'))
init_data = init_knowledge_base()
knowledge_base_init()
gl.set_value('INIT_DATA', init_data)
return product
@app.route('/getProfile', methods=['GET', 'POST'])
def get_profile():
ent = request.args.get('ent')
if ent is None:
return 'Get a None entity!'
if ent != '全部':
ent_detail = build_attr_tree(ent, 15)
file = open('static/result_json/' + gl.get_value('UPLOAD_FILE_PATH', None) + '_' + ent + '.js', 'wb')
file.write(
('data[\'' + ent + '\']=\'' + json.dumps(ent_detail) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
gl.set_value('STATE', 'free')
return 'succeed'
@app.route('/analysis', methods=['GET', 'POST'])
def analysis():
ent = None
type = request.args.get('type')
if type is None:
type = 'views'
if type == 'views':
ent = '全部'
gl.set_value('STATE', 'free')
elif type == 'details':
ent = request.args.get('name')
ent_detail = build_attr_tree(ent, 15)
file = open('static/result_json/' + gl.get_value('UPLOAD_FILE_PATH', None) + '_' + ent + '.js', 'wb')
file.write(
('data[\'' + ent + '\']=\'' + json.dumps(ent_detail) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
gl.set_value('STATE', 'free')
form = AnalysisForm()
input_text = None
upload_error = None
single_results = None
state_list = None
keywords = None
entity_pair = gl.get_value('ENTITY_PAIR', None)
if entity_pair is None:
knowledge_base_init()
entity_pair = gl.get_value('ENTITY_PAIR')
entity_level = gl.get_value('ENTITY_LEVEL')
if form.is_submitted():
if form.submit_text.data:
input_text = form.text.data
# sentiments, single_pairs = analysis_comment(text=input_text, debug=True, **init_data)
sentiments, state_list = analysis_comment(text=input_text,
init_data=gl.get_value('INIT_DATA'),
unlabeled_text=gl.get_value('UNLABELED_TEXT'))
# state_list = list(set([tuple(t) for t in state_list]))
if gl.get_value('STDOUT', False):
for state in state_list:
print(state.this_entity_name + ' ' + state.this_attribute_name + ' ' + state.this_va + ' ' + str(
state.this_score) + ' ' + state.text + ' ' + str(state.confidence))
input_text = form.text.data
single_results = single_analysis_results(state_list, entity_pair, entity_level)
elif form.submit_file.data:
try:
filename = texts.save(form.file.data)
gl.set_value('UPLOAD_FILE_PATH', filename)
file_url = texts.url(filename)
print(file_url)
gl.set_value('PROGRESS', 0)
gl.set_value('STATE', 'busy')
download_filepath = gen_summary(filename=filename, thread_num=thread_num)
# word_cloud_generate(filename)
profile_tree = multi_analysis_result(entity_pair)
ent_polar_include_children = dict()
compute_polar_include_children(profile_tree, ent_polar_include_children)
gl.set_value('PROFILE_TREE', profile_tree)
gl.set_value('ENT_POLAR_INCLUDE_CHILDREN', ent_polar_include_children)
summary, keywords = brief_summary()
print(summary)
gl.set_value('KEYWORDS', keywords)
word_cloud_generate()
gl.set_value('DOWNLOAD_FILE_PATH', 'downloads/' + download_filepath)
file = open('static/result_json/' + filename + '.js', 'wb')
file.write(('var data=new Array();\ndata[\'全部\']=\'' + json.dumps(profile_tree) + '\';').replace('\\"',
'\\\\"').encode(
'utf-8'))
file.write(
('\ndata[\'polars_include_children\']=\'' + json.dumps(ent_polar_include_children) + '\';').replace(
'\\"', '\\\\"').encode('utf-8'))
file.close()
ent = None
except UploadNotAllowed as una:
upload_error = 'error!'
print(una)
unlabeled_text = gl.get_value('UNLABELED_TEXT', [])
if (len(unlabeled_text) > 100):
flush_unlabeled_text()
return render_template('review_analysis.html', form=form,
input_text=input_text,
upload_error=upload_error,
download_filepath=gl.get_value('DOWNLOAD_FILE_PATH'),
single_results=single_results, state_list=state_list,
upload_file_name=gl.get_value('UPLOAD_FILE_PATH'),
profile_tree=gl.get_value('PROFILE_TREE'),
ent=ent,
product=gl.get_value('PRODUCT', '汽车'),
keywords=gl.get_value('KEYWORDS'))
from Fine_grained.Src.Fine_grained.STATE import State
@app.route('/test', methods=['GET', 'POST'])
def test():
S1 = State()
S1.text = "just a test"
S2 = State()
S2.text = "another test"
list = []
list.append(S1)
list.append(S2)
return render_template('test.html', list=list)
@app.route('/get_progress', methods=['GET', 'POST'])
def get_progress():
opt = request.args.get('opt')
if opt == 'reset':
gl.set_value('STATE', 'free')
gl.set_value('PROGRESS', 0)
return gl.get_value('STATE') + '-' + str(gl.get_value('PROGRESS'))
@app.route('/views', methods=['GET', 'POST'])
def menu():
form = AnalysisForm()
return render_template('review_analysis.html', form=form,
download_filepath=gl.get_value('DOWNLOAD_FILE_PATH', None),
upload_file_name=gl.get_value('UPLOAD_FILE_PATH', None), ent='全部')
@app.route('/details', methods=['GET', 'POST'])
def detail():
form = AnalysisForm()
ent = request.args.get('name')
ent_detail = build_attr_tree(ent, 15)
file = open('static/result_json/' + gl.get_value('UPLOAD_FILE_PATH', None) + '_' + ent + '.js', 'wb')
file.write(('data[\'' + ent + '\']=\'' + json.dumps(ent_detail) + '\';').replace('\\"', '\\\\"').encode('utf-8'))
file.close()
return render_template('review_analysis.html', form=form,
download_filepath=gl.get_value('DOWNLOAD_FILE_PATH', None),
upload_file_name=gl.get_value('UPLOAD_FILE_PATH', None), ent=ent)
def build_attr_tree(ent, sample_num):
ent_attr_polar = gl.get_value('ENT_ATTR_POLAR')
ent_attr_text = gl.get_value('ENT_ATTR_TEXT')
ent_polar = gl.get_value('ENT_POLAR')
root = {'name': ent, 'id': 1, 'pos': ent_polar[ent][0], 'neu': ent_polar[ent][1], 'neg': ent_polar[ent][2],
'child': [], 'type': 'entity'}
id = 1
for ent_attr, polar in ent_attr_polar.items():
if ent != ent_attr.split('-')[0]:
continue
attr = ent_attr.split('-')[1]
attr_node = {'name': attr, 'id': id + 1, 'pos': ent_attr_polar[ent_attr][0], 'neu': ent_attr_polar[ent_attr][1],
'neg': ent_attr_polar[ent_attr][2], 'child': [], 'type': 'attr'}
pos_node = {'name': '正向评价', 'id': id + 2, 'pos': ent_attr_polar[ent_attr][0],
'neu': ent_attr_polar[ent_attr][1], 'neg': ent_attr_polar[ent_attr][2], 'child': [],
'type': 'sentiment_node'}
neu_node = {'name': '中性评价', 'id': id + 3, 'pos': ent_attr_polar[ent_attr][0],
'neu': ent_attr_polar[ent_attr][1], 'neg': ent_attr_polar[ent_attr][2], 'child': [],
'type': 'sentiment_node'}
neg_node = {'name': '负向评价', 'id': id + 4, 'pos': ent_attr_polar[ent_attr][0],
'neu': ent_attr_polar[ent_attr][1], 'neg': ent_attr_polar[ent_attr][2], 'child': [],
'type': 'sentiment_node'}
id = id + 4
pos_sentence = random.sample(ent_attr_text[ent_attr][0], min(sample_num, len(ent_attr_text[ent_attr][0])))
for sentence in pos_sentence:
id = id + 1
node = {'name': sentence, 'id': id, 'pos': 0, 'neu': 0, 'neg': 0, 'child': [], 'type': 'sentence'}
pos_node['child'].append(node)
# pos_node['child'] = random.sample(pos_node['child'], min(sample_num, len(pos_node['child'])))
neu_sentence = random.sample(ent_attr_text[ent_attr][1], min(sample_num, len(ent_attr_text[ent_attr][1])))
for sentence in neu_sentence:
id = id + 1
node = {'name': sentence, 'id': id, 'pos': 0, 'neu': 0, 'neg': 0, 'child': [], 'type': 'sentence'}
neu_node['child'].append(node)
# neu_node['child'] = random.sample(neu_node['child'], min(sample_num, len(neu_node['child'])))
neg_sentence = random.sample(ent_attr_text[ent_attr][2], min(sample_num, len(ent_attr_text[ent_attr][2])))
for sentence in neg_sentence:
id = id + 1
node = {'name': sentence, 'id': id, 'pos': 0, 'neu': 0, 'neg': 0, 'child': [], 'type': 'sentence'}
neg_node['child'].append(node)
# neg_node['child'] = random.sample(neg_node['child'], min(sample_num, len(neg_node['child'])))
attr_node['child'].append(pos_node)
attr_node['child'].append(neu_node)
attr_node['child'].append(neg_node)
root['child'].append(attr_node)
return root
class AnalysisForm(FlaskForm):
text = TextAreaField('请输入待分析的单条评论：')
submit_text = SubmitField('提交')
file = FileField('请上传待分析的批量评论文件：')
submit_file = SubmitField('提交')
class TrainingForm(FlaskForm):
text = TextAreaField('请输入待分析的文本：')
submit_text = SubmitField('提交')
# text1 = TextAreaField('请输入训练用评论语料库：')
# submit_text1 = SubmitField('提交')
file = FileField('请上传待分析语料库文件：')
submit_file = SubmitField('提交')
aspect = FileField('aspect')
submit_aspect = SubmitField('显示产品构成')
# opinion=FileField('opinion')
# file_aspect = FileField('请上传实体属性关系文件：')
# submit_file_aspect = SubmitField('提交')
#
# file_entity_synonym = FileField('请上传实体本体库文件：')
# submit_file_entity_synonym = SubmitField('提交')
#
# file_attr_synonym = FileField('请上传属性本体库文件：')
# submit_file_attr_synonym = SubmitField('提交')
# file_opinion_pair = FileField('请上传情感库文件：')
# submit_file_opinion_pair=SubmitField('提交')
def read_aspect_file(file_path, entity_level=None):
flag = False
if entity_level is None:
flag = True
if flag:
entity_level = dict()
aspect_file = open(file_path, encoding='utf8')
try:
all_lines = aspect_file.readlines()
aspect_pair = []
for line in all_lines:
aspects = line.split()
if len(aspects) < 2:
continue
parent = aspects[0]
if parent not in entity_level:
entity_level[parent] = 1
for child in aspects[1:len(aspects)]:
aspect_pair.append([parent, child, entity_level[parent]])
if flag and (child not in entity_level):
entity_level[child] = entity_level[parent] + 1
finally:
aspect_file.close()
return aspect_pair, entity_level
def read_opinion_file(file_path):
opinion_file = open(file_path, encoding='utf8')
try:
all_lines = opinion_file.readlines()
opinion_pair = []
polarity = 2
for line in all_lines:
pair = line.split()
polarity = polarity % 3 - 1
if len(pair) < 2:
continue
aspect = pair[0]
for opinion in pair[1:len(pair)]:
opinion_pair.append([aspect, opinion, polarity])
finally:
opinion_file.close()
return opinion_pair
def read_subset_file(file_path):
subset_file = open(file_path, encoding='utf8')
subset = {'nodes': [], 'legend_entity': [], 'legend_attribute': [], 'legend_description': []}
try:
subset['nodes'] = subset_file.readline().split()
subset['legend_entity'] = subset_file.readline().split()
subset['legend_attribute'] = subset_file.readline().split()
subset['legend_description'] = subset_file.readline().split()
finally:
subset_file.close()
return subset
def single_analysis_function(input_text):
pair = [['汽车', '性能', '不错', 1], ['汽车', '价格', '能接受', 0], ['汽车', '外观', '大气', 1], ['内饰', '空间', '不是很大', -1]]
return pair
def multi_analysis_function(multi_review_path):
pair = [['汽车', '整体', 550, 306, 954], ['汽车', '造型', 156, 534, 962], ['汽车', '性能', 664, 875, 277],
['汽车', '空间', 761, 335, 699], ['汽车', '性价比', 801, 1, 200], ['汽车', '油耗', 611, 119, 284],
['汽车', '外观', 531, 533, 267], ['汽车', '动力', 313, 233, 314], ['汽车', '隔音', 322, 634, 385],
['汽车', '价格', 338, 773, 741], ['汽车', '配置', 984, 474, 948], ['汽车', '舒适性', 886, 680, 724],
['汽车', '质量', 82, 561, 747], ['汽车', '起步', 863, 986, 68], ['汽车', '款式', 130, 553, 241],
['汽车', '噪音', 811, 923, 916], ['汽车', '销量', 262, 950, 285], ['汽车', '功能', 297, 617, 829],
['汽车', '安全性', 392, 20, 461], ['汽车', '气味', 172, 732, 235], ['汽车', '品牌', 49, 356, 701],
['汽车', '设计', 616, 834, 224], ['汽车', '尺寸', 221, 498, 811], ['汽车', '材料', 803, 843, 275],
['汽车', '操控', 395, 961, 590], ['汽车', '表现', 280, 312, 527], ['汽车', '成本', 616, 159, 684],
['汽车', '档次', 654, 621, 716], ['汽车', '问题', 750, 206, 821], ['汽车', '速度', 531, 413, 794],
['汽车', '线条', 214, 242, 140], ['汽车', '声音', 678, 809, 209], ['汽车', '口碑', 551, 103, 485],
['汽车', '风格', 452, 415, 826], ['汽车', '重心', 117, 153, 676], ['汽车', '颜色', 507, 698, 744],
['汽车', '长度', 543, 367, 212], ['汽车', '宽度', 557, 658, 15], ['汽车', '重量', 136, 295, 848],
['汽车', '轴距', 629, 588, 145], ['汽车', '排量', 977, 737, 292], ['汽车', '转速', 140, 980, 685],
['汽车', '型号', 721, 550, 262], ['汽车', '视野', 523, 31, 399], ['汽车', '优点', 222, 419, 635],
['汽车', '缺点', 956, 860, 929], ['车头', '整体', 566, 931, 75], ['车头', '造型', 391, 91, 402],
['车头', '设计', 484, 495, 412], ['车头', '线条', 960, 322, 580], ['车头', '外观', 640, 416, 292],
['车头', '位置', 550, 676, 438], ['车身', '整体', 993, 373, 55], ['车身', '造型', 881, 927, 24],
['车身', '线条', 310, 206, 943], ['车身', '尺寸', 376, 887, 301], ['车身', '重量', 359, 981, 228],
['车身', '长度', 221, 973, 916], ['车身', '宽度', 161, 872, 583], ['车身', '高度', 855, 388, 385],
['车身', '质量', 686, 316, 449], ['车身', '外观', 23, 496, 604], ['车身', '颜色', 238, 767, 525],
['车身', '设计', 295, 839, 327], ['车身', '比例', 516, 258, 146], ['车身', '结构', 255, 676, 95],
['车身', '稳定性', 548, 958, 759], ['车尾', '整体', 585, 691, 913], ['车尾', '造型', 786, 178, 183],
['车门', '整体', 85, 842, 843], ['车门', '隔音', 959, 585, 683], ['车窗', '整体', 592, 897, 889],
['车轮', '整体', 628, 844, 195], ['车灯', '整体', 537, 535, 351], ['车灯', '造型', 963, 76, 481],
['车灯', '设计', 809, 472, 153], ['车灯', '亮度', 155, 436, 497], ['车灯', '外观', 92, 845, 717],
['车灯', '效果', 233, 998, 368], ['底盘', '整体', 252, 353, 404], ['底盘', '高度', 209, 146, 842],
['底盘', '质感', 752, 662, 645], ['底盘', '用料', 425, 544, 789], ['底盘', '设计', 808, 522, 830],
['底盘', '稳定性', 535, 829, 885], ['底盘', '隔音', 806, 477, 265], ['底盘', '减震', 318, 697, 251],
['发动机', '整体', 820, 83, 758], ['发动机', '声音', 884, 537, 950], ['发动机', '噪音', 188, 935, 212],
['发动机', '转速', 130, 992, 536], ['发动机', '启停', 583, 686, 340], ['发动机', '动力', 206, 417, 421],
['发动机', '技术', 752, 547, 876], ['发动机', '隔音', 632, 553, 775], ['发动机', '故障', 695, 587, 367],
['发动机', '排量', 330, 539, 43], ['发动机', '油耗', 84, 725, 580], ['发动机', '表现', 640, 642, 162],
['发动机', '功率', 775, 551, 185], ['发动机', '质量', 383, 548, 316], ['发动机', '温度', 621, 679, 555],
['发动机', '性能', 432, 403, 3], ['发动机', '马力', 36, 860, 371], ['发动机', '型号', 599, 785, 297],
['变速箱', '整体', 836, 685, 131], ['变速箱', '反应', 773, 713, 256], ['变速箱', '表现', 440, 821, 371],
['变速箱', '故障', 226, 567, 259], ['变速箱', '技术', 55, 760, 365], ['变速箱', '手感', 411, 209, 845],
['变速箱', '感觉', 834, 384, 723], ['变速箱', '平顺性', 835, 735, 924], ['油门', '整体', 183, 976, 539],
['油门', '反应', 198, 474, 670], ['油门', '行程', 386, 819, 681], ['油门', '控制', 560, 328, 329],
['油门', '位置', 84, 335, 495], ['油门', '感觉', 378, 193, 545], ['油门', '力度', 1, 799, 176],
['油门', '声音', 590, 326, 867], ['油门', '灵敏度', 771, 515, 638], ['刹车', '整体', 59, 700, 143],
['刹车', '反应', 21, 765, 945], ['刹车', '性能', 658, 854, 514], ['刹车', '力度', 721, 679, 653],
['刹车', '行程', 454, 865, 502], ['刹车', '位置', 524, 25, 433], ['刹车', '效果', 226, 21, 977],
['刹车', '问题', 717, 112, 575], ['刹车', '故障', 154, 547, 677], ['刹车', '声音', 900, 763, 697],
['刹车', '灵敏度', 699, 690, 144], ['离合器', '整体', 477, 525, 727], ['离合器', '行程', 435, 64, 817],
['离合器', '位置', 25, 899, 970], ['方向盘', '整体', 139, 939, 86], ['方向盘', '造型', 82, 605, 295],
['方向盘', '手感', 469, 107, 138], ['方向盘', '力度', 442, 652, 862], ['方向盘', '轻重', 752, 805, 774],
['方向盘', '尺寸', 812, 689, 933], ['方向盘', '设计', 252, 404, 139], ['方向盘', '操作', 295, 601, 307],
['方向盘', '材料', 783, 34, 645], ['方向盘', '控制', 365, 333, 691], ['方向盘', '功能', 101, 151, 819],
['仪表盘', '整体', 490, 149, 352], ['仪表盘', '设计', 397, 93, 309], ['仪表盘', '造型', 780, 51, 868],
['仪表盘', '背景', 693, 755, 683], ['仪表盘', '功能', 632, 426, 714], ['仪表盘', '亮度', 721, 507, 843],
['座椅', '整体', 422, 570, 4], ['座椅', '包裹', 881, 831, 942], ['座椅', '硬度', 683, 949, 705],
['座椅', '厚度', 323, 811, 414], ['座椅', '角度', 382, 650, 200], ['座椅', '宽度', 798, 661, 525],
['座椅', '设计', 41, 693, 635], ['座椅', '材料', 185, 908, 386], ['座椅', '感觉', 443, 766, 279],
['座椅', '手感', 216, 338, 328], ['座椅', '支撑', 831, 694, 171], ['座椅', '布局', 538, 518, 502],
['座椅', '位置', 379, 847, 921], ['座椅', '颜色', 606, 815, 520], ['座椅', '质量', 646, 819, 335],
['座椅', '舒适性', 880, 622, 335], ['空调', '整体', 867, 697, 966], ['空调', '效果', 965, 789, 720],
['空调', '动力', 747, 990, 860], ['空调', '油耗', 8, 350, 851], ['空调', '操作', 601, 806, 366],
['空调', '温度', 167, 985, 458], ['空调', '声音', 865, 877, 528], ['空调', '噪音', 5, 700, 690],
['空调', '感觉', 382, 893, 982], ['空调', '功能', 378, 860, 486], ['空调', '性能', 223, 880, 848],
['雷达', '整体', 867, 696, 628], ['雷达', '声音', 979, 765, 464], ['音响', '整体', 981, 33, 312],
['音响', '效果', 183, 158, 999], ['音响', '音质', 724, 214, 622], ['音响', '声音', 385, 934, 807],
['音响', '质量', 395, 905, 681], ['音响', '感觉', 121, 448, 391], ['音响', '音量', 528, 609, 877],
['内饰', '整体', 847, 825, 898], ['内饰', '材料', 295, 899, 676], ['内饰', '设计', 596, 541, 489],
['内饰', '风格', 578, 608, 532], ['内饰', '颜色', 295, 481, 454], ['内饰', '配置', 851, 597, 706],
['内饰', '质感', 751, 464, 602], ['内饰', '布局', 316, 508, 750], ['内饰', '造型', 248, 244, 131],
['内饰', '味道', 913, 707, 755], ['内饰', '档次', 546, 725, 354],
['内饰', '手感', 384, 139, 357], ['内饰', '外观', 375, 334, 35], ['内饰', '气味', 11, 249, 263],
['导航', '整体', 461, 26, 545], ['导航', '功能', 274, 465, 509], ['导航', '版本', 445, 356, 518],
['导航', '反应', 449, 518, 706], ['导航', '声音', 760, 471, 762], ['喇叭', '整体', 915, 518, 537],
['喇叭', '声音', 32, 939, 746], ['喇叭', '音质', 975, 170, 634], ['喇叭', '效果', 586, 40, 320],
['喇叭', '音量', 688, 353, 869], ['后备箱', '整体', 160, 260, 853], ['后备箱', '空间', 57, 27, 217],
['后备箱', '容量', 84, 314, 206], ['后备箱', '设计', 878, 81, 593], ['后备箱', '尺寸', 988, 399, 836],
['后备箱', '隔音', 869, 428, 406], ['后备箱', '长度', 6, 241, 138], ['后备箱', '深度', 353, 377, 370],
['后备箱', '宽度', 864, 96, 845], ['气囊', '整体', 363, 35, 566], ['气囊', '故障', 56, 55, 945],
['天窗', '整体', 72, 654, 348], ['天窗', '视野', 766, 660, 547], ['天窗', '面积', 663, 80, 184],
['天窗', '设计', 484, 797, 242], ['引擎盖', '整体', 219, 717, 487], ['引擎盖', '隔音', 463, 714, 529],
['引擎盖', '缝隙', 743, 117, 330], ['铰链', '整体', 902, 596, 493], ['储物格', '整体', 735, 688, 243],
['储物格', '设计', 854, 754, 194], ['储物格', '空间', 696, 425, 133], ['车门开关', '整体', 264, 948, 688],
['车门开关', '', 883, 525, 365], ['车门开关', '位置', 137, 598, 710], ['玻璃', '整体', 485, 482, 890],
['玻璃', '隔音', 630, 621, 140], ['玻璃', '面积', 104, 638, 771], ['车窗开关', '整体', 732, 745, 205],
['车窗开关', '', 54, 215, 682], ['车窗开关', '位置', 512, 585, 241], ['车窗按钮', '整体', 991, 819, 991],
['升降机', '整体', 296, 496, 735], ['轮胎', '整体', 3, 185, 451], ['轮胎', '噪音', 403, 421, 483],
['轮胎', '规格', 530, 59, 552], ['轮胎', '气压', 608, 274, 378], ['轮胎', '尺寸', 643, 324, 406],
['轮胎', '声音', 511, 170, 573], ['轮胎', '宽度', 977, 746, 876], ['轮胎', '问题', 720, 131, 284],
['轮胎', '花纹', 319, 494, 397], ['轮胎', '温度', 698, 284, 114], ['轮胎', '性能', 791, 273, 863],
['轮胎', '型号', 387, 707, 566], ['轮毂', '整体', 156, 808, 498], ['轮毂', '造型', 208, 539, 51],
['轮毂', '样式', 921, 185, 198], ['轮毂', '尺寸', 573, 105, 569], ['大灯', '整体', 131, 437, 522],
['大灯', '亮度', 968, 180, 898], ['大灯', '设计', 492, 724, 698], ['大灯', '造型', 695, 155, 284],
['大灯', '配置', 385, 532, 441], ['大灯', '接缝', 986, 435, 673], ['大灯', '外观', 519, 932, 607],
['大灯', '高度', 182, 404, 532], ['大灯', '位置', 43, 851, 880], ['大灯', '功能', 790, 652, 504],
['尾灯', '整体', 677, 359, 674], ['尾灯', '设计', 980, 865, 956], ['尾灯', '造型', 579, 761, 789],
['尾灯', '辨识度', 747, 829, 922], ['雾灯', '整体', 445, 940, 139], ['刹车灯', '整体', 784, 122, 890],
['皮带', '整体', 779, 445, 270], ['隔音板', '整体', 729, 155, 106], ['齿轮', '整体', 56, 659, 381],
['油门踏板', '整体', 982, 754, 794], ['油门踏板', '行程', 788, 403, 173], ['油门踏板', '位置', 677, 367, 601],
['刹车踏板', '整体', 597, 343, 942], ['刹车踏板', '行程', 622, 308, 152], ['刹车踏板', '位置', 470, 693, 620],
['离合器踏板', '整体', 403, 253, 422], ['离合器踏板', '行程', 472, 154, 968], ['离合器踏板', '位置', 138, 99, 534],
['椅面', '整体', 408, 806, 944], ['椅背', '整体', 289, 509, 938], ['椅背', '角度', 4, 253, 751],
['座垫', '整体', 82, 62, 942], ['安全带', '整体', 310, 589, 783], ['皮套', '整体', 468, 400, 510],
['滤芯', '整体', 501, 555, 66], ['压缩机', '整体', 681, 636, 508], ['压缩机', '噪音', 359, 296, 59],
['空调旋钮', '整体', 607, 184, 490], ['空调旋钮', '手感', 402, 551, 336], ['空调旋钮', '位置', 786, 265, 256],
['空调旋钮', '设计', 186, 555, 504], ['空调按钮', '整体', 523, 244, 992], ['空调开关', '整体', 313, 828, 360],
['空调开关', '位置', 510, 581, 783], ['排水孔', '整体', 607, 848, 127], ['遮阳帘', '整体', 585, 40, 659]]
return pair
def multi_analysis_result(entity_pair):
ent_attr_polar = gl.get_value('ENT_ATTR_POLAR', None)
ent_attr_text = gl.get_value('ENT_ATTR_TEXT', None)
ent_polar = dict()
ent_text = dict()
attr_polar = dict()
attr_text = dict()
ent_attribute = dict()
for ent_attr, polar in ent_attr_polar.items():
ent = ent_attr.split('-')[0]
attr = ent_attr.split('-')[1]
attrs = ent_attribute.setdefault(ent, [])
if attr not in attrs:
attrs.append(attr)
ent_attribute[ent] = attrs
polars = ent_polar.setdefault(ent, [0, 0, 0])
attr_polars = attr_polar.setdefault(attr, [0, 0, 0])
polars = (np.array(polars) + np.array(polar)).tolist()
attr_polars = (np.array(attr_polars) + np.array(polar)).tolist()
ent_polar[ent] = polars
attr_polar[attr] = attr_polars
attr_texts = attr_text.setdefault(attr, [[], [], []])
txts = ent_attr_text[ent_attr]
attr_texts[0].extend(txts[0])
attr_texts[1].extend(txts[1])
attr_texts[2].extend(txts[2])
gl.set_value('ATTR_POLAR', attr_polar)
gl.set_value('ATTR_TEXT', attr_text)
for ent, _ in ent_polar.items():
txts = ent_text.setdefault(ent, [[], [], []])
general_txt = ent_attr_text.setdefault(ent + '-' + '整体', [[], [], []])
for i in range(3):
txts[i] = txts[i] + general_txt[i]
if not txts[i]:
for attr in ent_attribute[ent]:
txt = ent_attr_text[ent + '-' + attr][i]
if txt:
txts[i] = txts[i] + txt
break
ent_text[ent] = txts
product = gl.get_value('PRODUCT', '汽车')
root = {'name': product, 'id': 1, 'pos': ent_polar[product][0], 'neu': ent_polar[product][1],
'neg': ent_polar[product][2],
'pos_sentence': ' || '.join(ent_text[product][0]), 'neu_sentence': ' || '.join(ent_text[product][1]),
'neg_sentence': ' || '.join(ent_text[product][2]), 'child': [], 'type': 'entity'}
id = 1
_ = build_tree(entity_pair, root, ent_text, id, ent_polar)
gl.set_value('ENT_POLAR', ent_polar)
return root
def brief_summary():
summary = ''
# words = 0
# review_num = 0
# try:
# with open('./uploads/' + gl.get_value('UPLOAD_FILE_PATH'), 'r', encoding='utf8') as fr:
# for line in fr:
# words += len(line.strip())
# review_num += 1
# except:
# pass
# if review_num == 0:
# review_num = 1
review_num = gl.get_value('REVIEW_NUM', 1)
words = gl.get_value('WORDS_NUM', 0)
summary += '上传的文件中共有%d条评论,平均每条评论有%d个字\n' % (review_num, words / review_num)
product = gl.get_value('PRODUCT')
ent_polar_include_children = gl.get_value('ENT_POLAR_INCLUDE_CHILDREN')
all_polars = ent_polar_include_children[product]
summary += '涉及情感倾向的评论片段共有%d个，其中正面的占%.1f%%，中性的占%.1f%%，负面的占%.1f%%\n' % (
sum(all_polars), 100 * all_polars[0] / sum(all_polars), 100 * all_polars[1] / sum(all_polars),
100 * all_polars[2] / sum(all_polars))
ent_polar = gl.get_value('ENT_POLAR')
best_entity = None
worst_entity = None
best_ent_polars = [0, 0, 0]
worst_ent_polars = [0, 0, 0]
for ent, polars in ent_polar.items():
if ent == product:
continue
positive = polars[0]
negative = polars[2]
if positive > best_ent_polars[0] or (positive == best_ent_polars[0] and sum(polars) < sum(best_ent_polars)):
best_entity = ent
best_ent_polars = polars
if negative > worst_ent_polars[2] or (negative == worst_ent_polars[2] and sum(polars) < sum(worst_ent_polars)):
worst_entity = ent
worst_ent_polars = polars
if best_entity is not None and worst_entity is not None:
summary += '其中，最受好评的部分是%s，最受诟病的部分是%s\n' % (best_entity, worst_entity)
attr_description = gl.get_value('ATTR_DESCRIPTION')
best_attr = None
worst_attr = None
best_num = 0
worst_num = 0
top_description_best = None
top_description_worst = None
best_attr_polars = [0, 0, 0]
worst_attr_polars = [0, 0, 0]
for attr, descriptions in attr_description.items():
polars = [len(descriptions[0]), len(descriptions[1]), len(descriptions[2])]
positive = polars[0]
negative = polars[2]
if positive > best_attr_polars[0] or (positive == best_attr_polars[0] and sum(polars) < sum(best_attr_polars)):
best_attr = attr
best_attr_polars = polars
if negative > worst_attr_polars[2] or (
negative == worst_attr_polars[2] and sum(polars) < sum(worst_attr_polars)):
worst_attr = attr
worst_attr_polars = polars
if best_attr is not None and worst_attr is not None:
best_descriptions = attr_description[best_attr][0]
worst_descriptions = attr_description[worst_attr][2]
from collections import Counter
best = Counter(best_descriptions).most_common(1)
worst = Counter(worst_descriptions).most_common(1)
best_num = best[0][1]
top_description_best = best[0][0]
worst_num = worst[0][1]
top_description_worst = worst[0][0]
summary += '大家对%s最为满意，有%d人认为它%s；同时，有%d人吐槽%s%s\n' % (
best_attr, best_num, top_description_best, worst_num, worst_attr, top_description_worst)