Testing fisher

[David Relson]

pi,

An interesting table.  Your best value of min_dev seems to depend on 
whether a non-zero F-P is acceptable to you.

min_dev=0.2 gives 184 F-N and 0 F-P, while min_dev of 0.015, 0.02, and 
0.025 give essentially identical results of 89-92 F-N and 1 F-P.  (I've 
extracted the min_dev lines I mention.)

    fisher-2        0.20          0.60     4221   184   15140  0

    fisher-2        0.025         0.60     4262    89   15251  1
    fisher-2        0.02          0.60     4297    92   15362  1
    fisher-2        0.015         0.60     4295    92   15361  1

So, having done your testing, what values are you actually using?

David

P.S. "Total" is the better column name.

At 09:16 AM 1/28/03, Boris 'pi' Piwinger wrote:

>David Relson wrote:
>
>Note: The numbers give the total size of the testbase, not
>the right detections.
>
>algorithm    min_def    spam_cutoff    test.spam    test.ham
>                                        total  F-N   total  F-P
>fisher-2        0.10          0.95     4186   364   15140  1
>fisher-2        0.25          0.60     4335   191   15362  0
>fisher-2        0.20          0.60     4221   184   15140  0
>fisher-2        0.15          0.60     4237   170   15251  0
>fisher-2        0.10          0.60     4221   139   15140  1
>fisher-2        0.075         0.60     4237   132   15251  1
>fisher-2        0.05          0.60     4237   116   15251  1
>fisher-2        0.035         0.60     4262   101   15251  1
>fisher-2        0.025         0.60     4262    89   15251  1
>fisher-2        0.02          0.60     4297    92   15362  1
>fisher-2        0.015         0.60     4295    92   15361  1
>fisher-2        0.00          0.60     4221   140   15140  1
>
>pi