Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

high false positive with Gmail

We've talked about Gmail's spam filtering on this list before, so in 
that light I'll mention that I just discovered I'm experiencing a fairly 
high false positive rate.

Several months back I switched over a bunch of mailing list 
subscriptions to my Gmail account on the theory that I could make use of 
Gmail's search tools to search through the archives (though I've yet to 
make use of that). This seemed to work fine, and I haven't used the web 
UI in months, instead downloading the messages via POP3 to Thunderbird.

Today a rare spam made it through to my inbox, so I logged in to the web 
UI to tag it as spam, and while there decided to check the spam folder. 
I was troubled by finding 3 or 4 messages incorrectly marked as spam 
just on the first screen of 50 messages. After reviewing all 333 
messages, 27 were incorrectly categorized as spam, or 8%. That's not so hot.

(The 8% is a bit misleading though, as a false positive rate would 
normally be stated as a percentage of the overall mail volume. 
Theoretically the search "in:anywhere after:2007/6/18 before:2007/8/19" 
should return a count covering the same range as the spam folder, but 
Gmail reports only 80 matches, yet perpetually boosts that number as you 
page through the results. Using my downloaded message archives, I'd 
estimate the number of messages to be around 5000 for that time period, 
so the false positive rate is more like 0.5%.)

I tagged them all as not spam, but as I understand it, that will only 
whitelist the specific senders, and not necessarily solve the long term 
problem. It seems Gmail has become a bit too aggressive in their spam 

Hopefully someday they'll add support for IMAP, which will make it much 
easier to keep tabs on the spam folder.


Tom Metro
Venture Logic, Newton, MA, USA
"Enterprise solutions through open source."
Professional Profile:

This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!

Boston Linux & Unix /