Hello Anthony, On 09.05.2012 13:13, P.V.Anthony wrote: > On 05/09/2012 06:02 PM, Stevan Bajić wrote: > >> You should! Creating a MERGED group that you use globally (aka global >> merged group) will allow you to create one big group and share that data >> with all the users. Allow me to explain (just an artificial example to >> illustrate the benefit for you): >> global merged group: 1'000'000 spam messages trained / 1'000'000 >> innocent messages trained -> 500MB data >> user 1: using all data from 'global merged group' + his own data: maybe >> 100 spam messages / 100 innocent messages -> 10 MB data >> user 2: using all data from 'global merged group' + his own data: maybe >> 10 spam messages / 10 innocent messages -> 1 MB data >> user 3: using all data from 'global merged group' + his own data: maybe >> 100 spam messages / 100 innocent messages -> 10 MB data >> user 4: using all data from 'global merged group' + his own data: maybe >> 10 spam messages / 10 innocent messages -> 1 MB data >> user n: using all data from 'global merged group' + his own data: maybe >> 100 spam messages / 100 innocent messages -> 10 MB data >> >> So the more users you have the less data per user is used since they all >> share the data from the globally merged user. For your setup I would >> strongly suggest to create that globally merged group and train it with >> dspam_train. Search the mailing list for details how to do that. In the >> past two to three weeks I have explained one approach of how to do it. > Need advice on groups. My users do not teach their dspam. I have to > repeat this again, my users do not teach their dspam. > > Currently I am using my account as GLOBAL GROUPS. Here is my group > setting in /home/dspam/group > groupGlobal:classification:*myacco...@example.com.sg > > I train my account whenever there is an error. I am using TOE. Here is > my dspam_stats. > > myacco...@example.com.sg: > TP True Positives: 18152 > TN True Negatives: 59943 > FP False Positives: 233 > FN False Negatives: 1216 > SC Spam Corpusfed: 0 > NC Nonspam Corpusfed: 0 > TL Training Left: 0 > SHR Spam Hit Rate 93.72% > HSR Ham Strike Rate: 0.39% > PPV Positive predictive value: 98.73% > OCA Overall Accuracy: 98.18% > > Is this the best type of group to use when my users do not train their > dspam? Or should I use merged group? The sad part is my users do not > train dspam. I love global merged groups but in your case the best group would probably be a global shared group. Are the users on your system having all +/- the same mail behaviour? I mean: do they get +/- the same type of mail and would they classify +/- the mail the same way? If the answer to this is yes then you should use a global shared group. If you have however a bunch of users having different behaviour then you could still go with a shared group by grouping the users with the common behaviour into one shared group and put the other either in no group at all or put them at least into a merged group and train from time to time the merged group with Spam (for example from http://untroubled.org/spam/) and with Ham from your system.
> Advice is appreciated. > > P.V.Anthony > > > > > -- Kind Regards from Switzerland, Stevan Bajić ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
