POPFile classified a message to bucket X, but the probabilities say it should be bucket Y

The classification POPFile displays on the history page may differ from the classification you can deduce from the probabilities table displayed below each message in the single message view. This can happen when you have reclassified a similar message after receiving the message in question.

Here's an example: You receive two messages that are similar. POPFile classifies both of them to your 'spam' bucket. After you have reclassified the first message to your 'personal' bucket, the probabilities for the second message will also change. Although the second message is still marked with its original 'spam' classification in POPFile's history, its probabilities will have changed, even up to the point where the classification should change as well.

Since classifying a message requires quite intensive computations, the cost of determining a new classification for each message in history each time a message is reclassified would be too high. Therefore, POPFile 0.22 will inform you about a changed classification at the top of the single message view.

 
faq/classvsscore.txt · Last modified: 2008/02/08 19:49 by 127.0.0.1

Should you find anything in the documentation that is incomplete, unclear, outdated or just plain wrong, please let us know and leave a note in the Documentation Forum.

Recent changes RSS feed Donate Driven by DokuWiki
The content of this wiki is protected by the GNU Fee Documentation License