Re: Some questions setting up latest CRM114
Bill Y <[email protected]>
| Newsgroups | gmane.mail.spam.crm114 |
|---|---|
| Message-ID | <[email protected]> |
From: Leonard Lin <[email protected]> > I'm finding that a pR threshold of +/- 5.0 is working better these > days, as it allows micro spams to be caught very easily (microspams > are spams with just a line of text and single URL as the payload) Thanks for the tip I'll take a look at the thresholds and see what . Do you know offhand about whether the X-CRM114 headers are all ignored when processing in mailreaver? I'm looking at the reaver/lib sources and see some mangle calls that I think might be doing the right thing but I'm not sure. I took a look as well- they appear to be in the outgoing /updated/confirmation path, not in the "mung before learning" path. So, the answer is "no, not ignored!" Also, and I guess I'll ask it here instead of crm114-discuss since there doesn't seem to be a lot of active responses there, but I'm interested in also using crm114 for classifying other datasets. * Has anyone used crm114 to successfully w larger (say, 1-10M item) datasets? Does it blow up either on insert (is it linear speed or does it slow down?) or for classifying? ~100K has been done at TREC. The datasets grow only when you have to train, so if you use TOE or SSTTT, the growth is logarithmic at worst and asymptotically limitend for most data. * And secondly, the hyperspace classifier seems really interesting because it sounds like (in theory) it would be good for different types of classification - I've seen some search results for n-way classification -- are there docs/examples for that? Yes, it does. Take a look at "gatlingsort" for an N-way classification. * Also, are there semantics for pulling "relatedness" (i.e. semantics for finding distance, or retrieving nearest neighbors)? At this point, no. The distance is "hidden" from the user, and the neighbors are stored only in a hashed form that (to an extent intentionally) obfuscates the original text. However, CLUMP will be able to do that for you. If you want to get into coding C, let me know and I'll give you the design briefing. - Bill Yerazunis ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace