Quote:
Originally Posted by stief dijital, sorry if this sounds too obvious, but marking the legit results as "not junk" is supposed to help such situations you describe, where the filter (baynesian, IIRC), can learn what you think is junk or not. |
I have done this for the files I tried downloading and found to be clean, but my searches are coming up with ~30 results marked as junk and maybe two or three as non-junk. In practice I am only going to download a single file from the list. I can mark that file as non-junk but the first problem with this is I am having to choose from the files marked as junk (which somewhat defeats the point of having a junk filter) and hope it was one of the incorrectly marked files and not a virus.
Secondly, if these files being incorrectly tagged
is a result of spammers deliberately tagging them wrong, then I would imagine they would not just mark the file once. They'd do it a couple of hundred times from different computers/proxies meaning that their tagging of the file would greatly outweigh mine and it would remain marked as junk.
I don't yet have the technical ability to help alter the algorithm, maybe once I have started my uni course I'll come back and see what I can do. At the moment I am just asking if anyone else has noticed the problem and if it's being worked on.