[Spambayes] Frequency distribution for wordinfo counts?

Kenny Pitt kennypitt at hotmail.com
Mon Feb 16 14:59:52 EST 2004


Tony Meyer wrote:
> Here's mine today; I'm using bigrams and 'url slurping' with a prefix.

And mine, with mine_received_headers and x-pick_apart_urls enabled.  All
other options are Outlook plugin defaults.

Db file size is 671,744 bytes

nham 79 nspam 92

ham counts
value  #times   cumm      %   cumm%
     0   7247   7247  42.95  42.95
     1   6694  13941  39.68  82.63
     2   1414  15355   8.38  91.01
     3    523  15878   3.10  94.11
     4    300  16178   1.78  95.89
     5    164  16342   0.97  96.86
     6     97  16439   0.57  97.43
     7     77  16516   0.46  97.89
     8     91  16607   0.54  98.43
     9     44  16651   0.26  98.69
    10     35  16686   0.21  98.90
    11     28  16714   0.17  99.06
    12     17  16731   0.10  99.16
    13     16  16747   0.09  99.26
    14     14  16761   0.08  99.34
    15     25  16786   0.15  99.49
    16      3  16789   0.02  99.51
    17      3  16792   0.02  99.53
    18     24  16816   0.14  99.67
    19      6  16822   0.04  99.70
    20      6  16828   0.04  99.74
    21      2  16830   0.01  99.75
    22      2  16832   0.01  99.76
    23      2  16834   0.01  99.77
    24      2  16836   0.01  99.79
    25      1  16837   0.01  99.79
    26      2  16839   0.01  99.80
    27      1  16840   0.01  99.81
    28      2  16842   0.01  99.82
    29      2  16844   0.01  99.83
    30      1  16845   0.01  99.84
    31      1  16846   0.01  99.85
    32      3  16849   0.02  99.86
    35      2  16851   0.01  99.88
    37      2  16853   0.01  99.89
    39      1  16854   0.01  99.89
    41      1  16855   0.01  99.90
    42      1  16856   0.01  99.91
    45      1  16857   0.01  99.91
    46      1  16858   0.01  99.92
    54      1  16859   0.01  99.92
    58      1  16860   0.01  99.93
    60      1  16861   0.01  99.93
    63      1  16862   0.01  99.94
    64      1  16863   0.01  99.95
    67      1  16864   0.01  99.95
    75      2  16866   0.01  99.96
    78      2  16868   0.01  99.98
    79      4  16872   0.02 100.00

spam counts
value  #times   cumm      %   cumm%
     0   7086   7086  42.00  42.00
     1   7612  14698  45.12  87.11
     2    961  15659   5.70  92.81
     3    523  16182   3.10  95.91
     4    231  16413   1.37  97.28
     5    109  16522   0.65  97.93
     6     61  16583   0.36  98.29
     7     47  16630   0.28  98.57
     8     38  16668   0.23  98.79
     9     28  16696   0.17  98.96
    10     17  16713   0.10  99.06
    11     26  16739   0.15  99.21
    12     12  16751   0.07  99.28
    13     17  16768   0.10  99.38
    14     10  16778   0.06  99.44
    15      8  16786   0.05  99.49
    16      6  16792   0.04  99.53
    17      9  16801   0.05  99.58
    18      6  16807   0.04  99.61
    19      7  16814   0.04  99.66
    20      7  16821   0.04  99.70
    23      5  16826   0.03  99.73
    24      2  16828   0.01  99.74
    25      2  16830   0.01  99.75
    26      1  16831   0.01  99.76
    27      1  16832   0.01  99.76
    28      1  16833   0.01  99.77
    29      2  16835   0.01  99.78
    30      1  16836   0.01  99.79
    31      1  16837   0.01  99.79
    32      1  16838   0.01  99.80
    34      2  16840   0.01  99.81
    35      1  16841   0.01  99.82
    40      3  16844   0.02  99.83
    42      2  16846   0.01  99.85
    47      1  16847   0.01  99.85
    48      2  16849   0.01  99.86
    49      2  16851   0.01  99.88
    53      1  16852   0.01  99.88
    54      1  16853   0.01  99.89
    61      1  16854   0.01  99.89
    62      1  16855   0.01  99.90
    63      1  16856   0.01  99.91
    64      1  16857   0.01  99.91
    67      1  16858   0.01  99.92
    68      1  16859   0.01  99.92
    69      2  16861   0.01  99.93
    73      1  16862   0.01  99.94
    75      1  16863   0.01  99.95
    82      1  16864   0.01  99.95
    85      1  16865   0.01  99.96
    86      1  16866   0.01  99.96
    88      2  16868   0.01  99.98
    89      1  16869   0.01  99.98
    90      1  16870   0.01  99.99
    91      1  16871   0.01  99.99
    92      1  16872   0.01 100.00

-- 
Kenny Pitt




More information about the Spambayes mailing list