0 votes

In your help on the spam filter, you say:

Both spam and ham folders should contain around 100-200 messages each
(the more the better).

Does that mean "the more the better" without any limits, or "the more the better up to around 200"?

In other words, if more than 200 emails are available for both good and bad emails - say there are 500, or 1000, or 2000 of each - should we use all emails available; or is there minimal value in using more than 200 and we should cap it at that number?

by (400 points)

1 Answer

0 votes

100 is minimum, 200 gives more or less good results. More will get you better results. Remember that both spam and ham should contain similar email count.

by (301k points)
...