|
|
|
|
|
 |
Thursday, September 19, 2002 |
|
More positive test results re my proposed improvements to Graham's spam filter. This is the only test so far to combine two of my proposals.
Also, I was curious about why so many spam projects are using home-grown techniques rather than the "naive Bayesian classification" approach that has been pursued fairly intensively in academia. I asked Tim Peters, founder of the spambayes project, and he gave his thoughts. |
|
From the spambayes mail list:
How do you get 60 rowdy, drunken Canadians out of a swimming pool on a hot summer day? "OK, guys, it's time to get out of the pool now..." |
 |
 |
|
 |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||