data_analysis.md

April 4, 2019 ยท View on GitHub

Data Analysis

  • Laptop
itemsize
training set2081
valid set232
test set638
word_vocab3595
char_vocab59
aspect1181
aspect_text_word_vocab981
aspect_text_char_vocab45
word_max_len83
< word_len = 0.9854
word_left_max_len70
< word_left_len = 0.97534
word_right_max_len78
< word_right_len = 0.98539
char_max_len465
< char_len = 0.98255
char_left_max_len365
< char_left_len = 0.98177
char_right_max_len400
< char_right_len = 0.981177
aspect_text_word_max_len8
< aspect_text_word_len = 0.996
aspect_text_char_max_len58
< aspect_text_char_len = 0.9928
  • resturant(term)
itemsize
training set3241
valid set361
test set1120
word_vocab4550
char_vocab58
aspect1528
aspect_text_word_vocab1407
aspect_text_char_vocab39
word_max_len79
< word_len = 0.97343
word_left_max_len72
< word_left_len = 0.98534
word_right_max_len72
< word_right_len = 0.98534
char_max_len358
< char_len = 0.98226
char_left_max_len344
< char_left_len = 0.98163
char_right_max_len326
< char_right_len = 0.981162
aspect_text_word_max_len21
< aspect_text_word_len = 0.996
aspect_text_char_max_len115
< aspect_text_char_len = 0.9932
  • resturant(category)
itemsize
training set3166
valid set352
test set973
word_vocab5175
char_vocab59
aspect5
word_max_len79
< word_len = 0.97135
char_max_len357
< char_len = 0.98189
  • twitter
itemsize
training set5623
valid set625
test set692
word_vocab13522
char_vocab116
aspect118
aspect_text_word_vocab176
aspect_text_char_vocab30
word_max_len73
< word_len = 0.99737
word_left_max_len39
< word_left_len = 0.98528
word_right_max_len67
< word_right_len = 0.99432
char_max_len188
< char_len = 0.985151
char_left_max_len156
< char_left_len = 0.98125
char_right_max_len164
< char_right_len = 0.981138
aspect_text_word_max_len3
< aspect_text_word_len = 0.992
aspect_text_char_max_len21
< aspect_text_char_len = 0.9916