 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Mac Forum / Applications / Eudora / February 2008Junk Filter Question
I have designated an email address to receive payment notification emails from paypal when ever a sale is made on a website... Unfortunately, the notifications keep getting caught in my Junk mailbox. I have set a filter to transfer any email from 'paypal.com' to a certain folder, and I have set another filter to transfer any email with the subject 'Notification of payment received' (which is always the subject header for this notices), as a back up / failsafe. Despite both filters (which have been dragged to the top of my filter list), these notification emails are still getting directed to my Junk mailbox. How can I change my filters to get these emails to be properly directed where I choose so that I don't run the risk of missing/losing them in my Junk folder? Thanks for any advice with this! jb Oooops! Forgot to mention I'm using Eudora 6.2.4 with OS 10.4.11... >Oooops! > >Forgot to mention I'm using Eudora 6.2.4 with OS 10.4.11... When such messages get directed to the Junk folder, have you been identifying them as "not junk?" Regards, Mike -- | Systems Specialist: CBE,MSE Michael T. Davis (Mike) | Departmental Networking/Computing http://www.ecr6.ohio-state.edu/~davism/ | The Ohio State University | 197 Watts, (614) 292-6928 > When such messages get directed to the Junk folder, have you been > identifying them as "not junk?" Somehow I'd not understood that the 'not junk'-ing of an email was remembered for future reference... I think this will resolve my problem. Many thanks! > > When such messages get directed to the Junk folder, have you been > > identifying them as "not junk?" > > Somehow I'd not understood that the 'not junk'-ing of an email was > remembered for future reference... I think this will resolve my > problem. To get maximum value from the spam filter, you need to train it. To do this, if a message is spam but Eudora does not detect it, you need to mark it as junk so Eudora will learn how to identify junk mail better. Similarly, if a message appears in the Junk folder but is not junk, you need to mark it as "not junk". Again, the spam filter will learn to recognise that messages like that are not spam. However, this a black-and-white distinction. Eudora's spam filter actually operates in shades of grey - it scores messages on a scale of 1 to 100. If a message's score is greater than a certain value, it is regarded as junk mail. In my case, the value is 40. Anything over 40 is junk, anything under 40 is ok. Now what if a good message comes in as 37? The spam filter has been very close to calling this junk. In fact, over time, similar messages may end up with scores over 40 unless you do something about it. (This I guess is an aspect of the algorithm, that lots of almost identical messages probably indicates spam.) So it is worthwhile marking good messages whose score is above 10 or 20 as "not junk" so that Eudora learns that those types of messages are really ok. Similarly, some messages in Junk are only slightly over 40, meaning that the spam filter had some doubts about whether they were spam. If they really are spam, it is worth telling Eudora that they are spam so as to remove any doubts. In summary, what this means is: 1. Periodically open your Junk folder and mark all the messages as junk (after checking that there are no false positives). 2. If any good e-mails have a score of 20 or more, mark them as "not junk". Using this method, I get pretty good results. According to Eudora's statistics, I received 21638 messages last year. Of those, there were 390 messages (1.8%) not identified correctly as junk (ie, I had to mark them as junk). There were 80 messages (0.3%) incorrectly identified as junk. The total number of junk messages was 1545, or about 7% of my messages. (It would be interesting to know what success rate other people are getting with the spam filter.) Cheers David ....... > (It would be interesting to know what success rate other people are > getting with the spam filter.) I am using Spamsieve with Eudora and it works pretty well. I started working with it in September last year and it seems to be trained well now. Since the last 3 months, 33.300 Spams have been sorted out and just 114 were sorted out wrong and had to be trained. That means 99.7% success rate. It should be better now after training these 114. >To get maximum value from the spam filter, you need to train it. To do >this, if a message is spam but Eudora does not detect it, you need to >mark it as junk so Eudora will learn how to identify junk mail better. >Similarly, if a message appears in the Junk folder but is not junk, you >need to mark it as "not junk". Again, the spam filter will learn to >recognise that messages like that are not spam. But many spam filters only get told "yes this is spam", or "no this isn't spam". I want them to know why. (Same thing with spell checkers - when I add a word to a dictionary, I'd like to have a pop up with it guessing its part of speech along with variations of that word such as plurals or change that word to a phrase). > But many spam filters only get told "yes this is spam", or "no this > isn't spam". I want them to know why. The whole point of heuristic learning (which I believe the spam filters use) is that they derive their own rules from the examples you give them. These rules can be very complicated. This avoids human programming, which is the source of so many errors.  SignaturePeter >> But many spam filters only get told "yes this is spam", or "no this >> isn't spam". I want them to know why. [quoted text clipped - 3 lines] >them. These rules can be very complicated. This avoids human >programming, which is the source of so many errors. That's fine - within my guidelines. It's like Amazon trying to guess what books I like from what books I have bought - sometimes it hits, sometimes it misses. But it can't read my mind. I don't want to program it - I just want to give it guidance. (Not to mention domains from which I want to see all e-mail). > I don't want to program it - I just want to give it guidance. You do that by marking mails as 'not junk', or leaving them in the junk. I'm not sure of the details as my ISP's spam filter is perfectly adequate for my purposes- added of course to the simple filtering rules I've made in Eudora. SignaturePeter >> I don't want to program it - I just want to give it guidance. > >You do that by marking mails as 'not junk', or leaving them in the junk. I guess I'm not very good at communicating. It does *not* know what criteria I use in determining junk. It has false positives and misses negatives. It could be improved by knowing what I want. > >> I don't want to program it - I just want to give it guidance. > > [quoted text clipped - 5 lines] > > It could be improved by knowing what I want. And you tell it what you want by correcting its behavior. If you find a false positive, correct it. If you find a negative, correct it. Over time, you should see a gradual increase in accuracy. SignatureNote: Please send all responses to the relevant news group. If you must contact me through e-mail, let me know when you send email to this address so that your email doesn't get eaten by my SPAM filter. JR > I guess I'm not very good at communicating. No, you made it perfectly clear. Provably I was the one who put it badly... > It does *not* know what > criteria I use in determining junk. But by marking mails 'junk' or 'not junk', you are indicating your criteria, and the program gradually learns them. It then applies those criteria- *your* criteria- to new incoming mail. That's how heuristics work. > It could be improved by knowing what I want. Indeed. Which is precisely what you are indicating when you go through the junk mailbox and sort the mail within it according to your criteria. SignaturePeter >> It could be improved by knowing what I want. > >Indeed. Which is precisely what you are indicating when you go through >the junk mailbox and sort the mail within it according to your criteria. It could be improved by knowing why I classify one e-mail as junk and another that would look to you to be very similar - not junk. > It could be improved by knowing why I classify one e-mail as junk and > another that would look to you to be very similar - not junk. Possible, but very unlikely. Heuristic algoriths, if properly written and regularly fed with info, are almost certainly cleverer than we are. SignaturePeter >> It could be improved by knowing why I classify one e-mail as junk and >> another that would look to you to be very similar - not junk. > >Possible, but very unlikely. Heuristic algoriths, if properly written >and regularly fed with info, are almost certainly cleverer than we are. I don't disagree. But I wasn't proposing a choice between the two, I was proposing an integration of two heads. > I don't disagree. But I wasn't proposing a choice between the two, I > was proposing an integration of two heads. In that case, you've already got it. Nothing stops you writing your own filters which would work in conjunction with Eudora's junk filter. Or am I wrong about that..? SignaturePeter > >> It could be improved by knowing what I want. > > [quoted text clipped - 3 lines] > It could be improved by knowing why I classify one e-mail as junk and > another that would look to you to be very similar - not junk. I doubt that. SignatureNote: Please send all responses to the relevant news group. If you must contact me through e-mail, let me know when you send email to this address so that your email doesn't get eaten by my SPAM filter. JR [removed non-Eudora group again -- what do they know about Eudora?] > It could be improved by knowing why I classify one e-mail as junk > and another that would look to you to be very similar - not junk. How do you propose to convey this "knowledge" to Eudora? For recognizing specific senders (real or forged) as "not junk," you can use the special "address book" bypass. You can take complete control by turning off automatic junking, and using your own filters exclusively. At any point that you have finished applying your own rules, if you wish to let the automatic junk scorer also have a say, you can add a filter to test its junk score. -- > > But many spam filters only get told "yes this is spam", or "no this > > isn't spam". I want them to know why. [quoted text clipped - 3 lines] > them. These rules can be very complicated. This avoids human > programming, which is the source of so many errors. That's not really how most training-based spam filters work. Most are based on a system known as "Naive Bayesian" classification. That approach isn't based on complex rules, but rather on a lot of very simple rules with scores derived statistically. The words in the message are tallied to a big list of word frequencies in spam and non-spam messages, and that list of relative frequencies is used as a lot of very simple scoring rules, with each rule testing the presence of a word and adding a score based on the frequency of the word in prior spam and non-spam. SignatureNow where did I hide that website... > > The whole point of heuristic learning (which I believe the spam filters > > use) is that they derive their own rules from the examples you give [quoted text clipped - 10 lines] > adding a score based on the frequency of the word in prior spam and > non-spam. I expressed it badly; the point is that this process, involving many simple criteria, is in itself a complicated one. SignaturePeter > > > The whole point of heuristic learning (which I believe the spam filters > > > use) is that they derive their own rules from the examples you give [quoted text clipped - 13 lines] > I expressed it badly; the point is that this process, involving many > simple criteria, is in itself a complicated one. Well, I guess so... But only if you consider probabilistic analysis complicated. <g,d,&r> SignatureNow where did I hide that website... [removed non-Eudora group] On Thu, 07 Feb 2008 20:44:18 -0600: > Periodically open your Junk folder and mark all the messages as junk > (after checking that there are no false positives). Is there any need to re-junk already junked mail? One could speculate whether re-junking (or re-non-junking) would any better train the algorithm, or whether in any cases it might prove counter-productive; might take a lot of research to find out. By the way, it appears that in the Windows version, it may not even be possible to re-junk what's already in Junk mailbox, nor un-junk what isn't even in Junk mailbox, which suggests discouragement of the practice. -- > Is there any need to re-junk already junked mail? > > One could speculate whether re-junking (or re-non-junking) > would any better train the algorithm, or whether in any cases > it might prove counter-productive; might take a lot of research to find out. I had a problem with mail being incorrectly junked due to my error in a bad filter. If I select the message in the Junk mail box and un-junk it, it gets transferred to the mailbox it should have gone into (the In box if no filter directs it elsewhere) but it retains the junk score of 100. If I again un-junk it, the Junk score is removed and all messages from that sender are selected and also have the junk score removed. SignatureKathy - If you're reading this in your web browser from Google or similar forum, NNTP "newsreaders" are a better way to access the content. <http://www.aptalaska.net/~kmorgan/how-it-works.html> Links to NNTP newsreaders at <http://www.newsreaders.com/> > [removed non-Eudora group] > [quoted text clipped - 8 lines] > would any better train the algorithm, or whether in any cases > it might prove counter-productive; might take a lot of research to find out. Not really. The Bayesian database used for SpamWatch is just plain text, and it takes a few seconds to confirm that explicity telling Eudora that a message is or is not junk changes the token counts for words in that message, whatever the prior classification status of the message. In principle, it is best for any Bayesian mail filtering system to be fed by correct classifications of every message it sees, but in practice that would make the whole thing conceptually worthless, so real filters like the one in Eudora usually use a combination of train-on-error and feedback auto-learning, but actual user confirmation of a classification is always more valuable as a basis for feeding the filter than using the filter's determination as a basis for auto-learning. > By the way, it appears that in the Windows version, > it may not even be possible to re-junk what's already in Junk mailbox, > nor un-junk what isn't even in Junk mailbox, Not so for the Mac version. > which suggests discouragement of the practice. Which would be unwise. I believe based on its behavior that the last version of the Bayesian filter in Mac Eudora has unspecified auto-learning thresholds for both spam and non-spam that are different from the classification threshold. Any similar filter that does any auto-learning should work that way, i.e. only automatically feed back the messages that are certainly spam or not spam, not the borderline messages. SignatureNow where did I hide that website... > Oooops! > > Forgot to mention I'm using Eudora 6.2.4 with OS 10.4.11... In that case, what, exactly, is marking the messages as junk. Is it your ISP or is it Eudora's ownn spam filter (which I never used because I refuse to pay for spam in anyway, and paying to manage spam is paying for it). SignatureBart: This is the worst day of my life. Homer: This is the worst day of your life SO FAR. On Wed, 06 Feb 2008 18:49:03 -0600: > I refuse to pay for spam anyway, > and paying to manage spam is paying for it. Is paying for pest-control the same as paying the pests? Stick to those principles, then, and let them invade :) -- [groups narrowed to Eudora only] > Despite filters, [PayPal] notification emails > are still getting directed to my Junk mailbox. IIRC, automatic junk filtering, when turned on, junks mails that score above the designated threshhold, before any user-specified filtering occurs; in other words, it pre-empts your filters. In this case, you have to either "un-junk" the mail yourself (perhaps better "training" your junk filter) or else turn off automatic filtering, and use an additional filter of your own to "junk" mail that has an above-threshhold junk score, after you have first done other filtering that you want to occur before junking. -- > [groups narrowed to Eudora only] > [quoted text clipped - 13 lines] > after you have first done other filtering > that you want to occur before junking. Another option is to put the necessary "paypal.com" addresses in an address book, and then select Settings->Junk Mail->Mail isn't junk if the sender is in an address book. The downside to this approach is, of course, that if someone fakes the paypal.com address, it *won't* get marked as junk. Roberta SignatureRoberta Millstein usenet@spamaway.rlm.net Remove "spamaway" to reply -- Posted via a free Usenet account from http://www.teranews.com > I have designated an email address to receive payment notification > emails from paypal when ever a sale is made on a website... [quoted text clipped - 18 lines] > > jb You've already adjusted the order of your rules. Next try adding a "stop processing rules" option to those filters. Using this option, the desired emails should land where expected and Eudora should leave them alone. The remaining rules would still be applied against messages that did not meet the criteria the first two filters (from paypal.com and the specific subject line). I don't use Eudora so don't know how or if this can be done with this application; however since this filtering option is available in every other email app that I've used I would be surprised if Eudora didn't have it too. SignatureSharon F In article <sharonfDELETE-1D9762.09570608022008@newsgroups.comcast.net>, > > I have designated an email address to receive payment notification > > emails from paypal when ever a sale is made on a website... [quoted text clipped - 21 lines] > You've already adjusted the order of your rules. Next try adding a "stop > processing rules" option to those filters. I don't know about Eudora, but with Mail.app, setting a "Stop evaluating rules" rule in any rule set prevents the automatic spam filter from working in all of them. Isaac > I don't know about Eudora, but with Mail.app, setting a "Stop evaluating > rules" rule in any rule set prevents the automatic spam filter from > working in all of them. Traditionally, "stop processing" or "stop evaluating" as Mail expresses it - applies on a message by message basis. If Mail is doing anything else, I would call that a bug. I use that option in my rules and my junk filtering seems to be trucking along as usual. I'm not saying you're wrong - just that I'm not seeing the same thing here. SignatureSharon F In article <sharonfDELETE-1C8ACA.14241708022008@newsgroups.comcast.net>, Sharon F <sharonfDELETE@THISmvps.org> wrote: > > I don't know about Eudora, but with Mail.app, setting a "Stop evaluating > > rules" rule in any rule set prevents the automatic spam filter from [quoted text clipped - 7 lines] > along as usual. I'm not saying you're wrong - just that I'm not seeing > the same thing here. The junk mail filter needs to be the first one. Anything matching a rule run before the spam filter, where that rule includes a "stop processing" flag, will by definition NOT be processed by the spam filter. It ain't rocket science. SignatureHelp improve usenet. Kill-file Google Groups. http://improve-usenet.org/ > I don't know about Eudora, but with Mail.app ... If you would like to see non-confusing answers that actually relate to and are valid for Eudora, please try trimming the "Newsgroups" of postings to just the desired Eudora group(s), thanks. -- > The junk mail filter needs to be the first one. Anything matching a > rule run before the spam filter, where that rule includes a "stop > processing" flag, will by definition NOT be processed by the spam > filter. > > It ain't rocket science. Since the spam filters are run first and it's trapping wanted mail, tweaking the spam filters or using "training" are other options. That's already been covered by other replies. If spam filtering came second why would stop processing be a problem? It's wanted mail - having those desired messages processed further by the spam filters is not necessary. Telling the application to "stop processing" these wanted messages and to move on to filtering other messages would be appropriate and helpful. I wasn't aware that spam filters ran first. Now I am. Thanks. SignatureSharon F > You've already adjusted the order of your rules. Next try adding a "stop > processing rules" option to those filters. Using this option, the [quoted text clipped - 7 lines] > other email app that I've used I would be surprised if Eudora didn't > have it too. Eudora does have the option, but the Junk filtering is done before any other filters are checked, so the message would already have gone to the Junk mailbox before getting to the rule telling it to stop processing. I think the only thing OP can do is add Paypal to his address book, as Roberta suggested. SignatureKathy - If you're reading this in your web browser from Google or similar forum, NNTP "newsreaders" are a better way to access the content. <http://www.aptalaska.net/~kmorgan/how-it-works.html> Links to NNTP newsreaders at <http://www.newsreaders.com/> > Eudora does have the option, but the Junk filtering is done before any > other filters are checked, so the message would already have gone to the > Junk mailbox before getting to the rule telling it to stop processing. > I think the only thing OP can do is add Paypal to his address book, as > Roberta suggested. I agree since junk is filtered first. SignatureSharon F |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.