Skip to content

Instantly share code, notes, and snippets.

@michaeltwofish
Created June 25, 2010 02:37
Show Gist options
  • Save michaeltwofish/452310 to your computer and use it in GitHub Desktop.
Save michaeltwofish/452310 to your computer and use it in GitHub Desktop.
> data
Length Method Naive.Bayes
1 500 default 0.9072
2 500 stopping 0.8788
3 500 stemming 0.9038
4 500 word counts 0.8992
5 500 normalise length 0.8747
6 500 count and normalise 0.9074
7 500 TF/IDF 0.8993
8 1000 default 0.9068
9 1000 stopping 0.8791
10 1000 stemming 0.9031
11 1000 word counts 0.8987
12 1000 normalise length 0.8817
13 1000 count and normalise 0.9115
14 1000 TF/IDF 0.8988
15 1500 default 0.9067
16 1500 stopping 0.8825
17 1500 stemming 0.9000
18 1500 word counts 0.8996
19 1500 normalise length 0.8886
20 1500 count and normalise 0.9126
21 1500 TF/IDF 0.8997
22 2000 default 0.9077
23 2000 stopping 0.8806
24 2000 stemming 0.9024
25 2000 word counts 0.9004
26 2000 normalise length 0.8884
27 2000 count and normalise 0.9132
28 2000 TF/IDF 0.9005
29 2500 default 0.9076
30 2500 stopping 0.8753
31 2500 stemming 0.9017
32 2500 word counts 0.8990
33 2500 normalise length 0.8861
34 2500 count and normalise 0.9130
35 2500 TF/IDF 0.8992
36 3000 default 0.9076
37 3000 stopping 0.8753
38 3000 stemming 0.9017
39 3000 word counts 0.8990
40 3000 normalise length 0.8849
41 3000 count and normalise 0.9136
42 3000 TF/IDF 0.8991
43 3500 default 0.9076
44 3500 stopping 0.8753
45 3500 stemming 0.9017
46 3500 word counts 0.8990
47 3500 normalise length 0.8853
48 3500 count and normalise 0.9136
49 3500 TF/IDF 0.8991
50 4000 default 0.9076
51 4000 stopping 0.8753
52 4000 stemming 0.9017
53 4000 word counts 0.8990
54 4000 normalise length 0.8853
55 4000 count and normalise 0.9142
56 4000 TF/IDF 0.8991
57 4500 default 0.9076
58 4500 stopping 0.8753
59 4500 stemming 0.9017
60 4500 word counts 0.8990
61 4500 normalise length 0.8853
62 4500 count and normalise 0.9142
63 4500 TF/IDF 0.8991
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment