-
Notifications
You must be signed in to change notification settings - Fork 0
/
answers.txt
executable file
·45 lines (35 loc) · 1.12 KB
/
answers.txt
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
Name: Wei Xu
Uniquename: weixu
/*********************************************************/
Accuracy of my Naive Bayes classifier:
1. Only Tokenization: 0.719388
2. Also remove stopwords(without stemming): 0.724490
3. Also stem the words(without stop words removing): 0.765306
4. Also remove the stopwords and stem the words: 0.714286
/*********************************************************/
The top 10 words with highest conditional probability in each class
(listed in reverse order)
1. In “lie” class
and 0.026999432785
she 0.0256381168463
to 0.0251843448667
is 0.0239364719229
i 0.021667612025
a 0.0178105501985
he 0.01622234827
the 0.0156551332955
me 0.0114577424844
her 0.0110039705048
2. In “True” class
and 0.0312850205454
i 0.0289503175196
we 0.024000747105
to 0.0195181172955
she 0.018023907359
the 0.0174635786328
is 0.0171834142697
my 0.0161561449384
me 0.0153156518491
a 0.015035487486
Discussion: Since we do not remove the stop words, it’s expected that the words with highest conditional probability will be these stop words which have high occurrence.
/*********************************************************/