|
|
|
||||||||||||||||||||||||||||
 |
|||||||||||||||||||||||||||||
 |
 |
||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
 |
My Vital
Statistics! |
||||||||||||||||||||||||||||
| Mailwasher
is a product of Firetrust, a
commercial package that offers multiple strategies for the effective
control of SPAM. |
|||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
 |
As an example
of what "BLScan and
Mailwasher does for
me", I attach some screen captures of my own vital statistics. |
*
BLScan * BLComp * BLConv * * BLScan * BLComp * BLConv * * BLScan * BLComp * BLConv * * BLScan * BLComp * BLConv |
|||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
|
The image below shows
the summary screen.
The Final Summary section shows that in the entire time I have been using Mailwasher, I have received ~22,500 emails with 87% being spam. The Spam, Legitimate Mail and Action Taken entries reflect a four week period from mid-September through mid-October 2004. |
|||||||||||||||||||||||||||||
| Graphic
1 |
 |
||||||||||||||||||||||||||||
The
second graphic below shows how much junk mail each spam defence
detected for the four week period. Looking at this
period we can see that the blacklisted addresses/domains is responsible
for ~79% of recognized spam. This is summarized in the following table.
|
|||||||||||||||||||||||||||||
| Graphic 2 |
 |
||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
| The third
graphic below shows how much junk mail each spam defence detected for
the four week period. As regards legitimate email, the same table is shown below. We see that 79% of legitimate email is from 'friendly' addresses, ~9% from filters that search for possible friendly incoming email and 12% is detected by bayesian filtering.
Note that the 'Restored'
figure is email deleted that I needed to recover from the trash. This
figure is the number of 'false positives' incurred by the Blacklist
Servers (such as SpamCop, FirstAlert!, SpamHaus etc).
|
|||||||||||||||||||||||||||||
| Graphic 3 |  |
||||||||||||||||||||||||||||
| * BLScan * BLComp * BLConv * |
|||||||||||||||||||||||||||||
| The final graphic below shows how much junk mail was detected (in red) versus the incoming mail (green) for the four week period. | |||||||||||||||||||||||||||||
| Graphic
4 |
 |
||||||||||||||||||||||||||||
| From
these graphs I hope you can see that the Blacklist absorbs the lion's
share of the credit for spam detection. I am not able to look at these
statistics over a greater time period (such as a year) but I have
observed some fluctuation in the figures month by month. I have no data
to present for this but I estimate my overall blacklist contribution is
~85% on average. |
|||||||||||||||||||||||||||||
| Back to top | |||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
|
|