Sorry Chris, I am snowed under for the rest of the week. Big sim analysis at work and finishing revisions on the book at night.
If you can ping me next week, I should be able to take a look again. My basic expectation is that you don't have anything going wrong. If you want to see positive only terms, you might try tweaking the ModelDissector to sort on value descending rather than absolute value descending. Coefficient values of 0.1 are about right for a model like this and with 20 newsgroups, it isn't so surprising to see lots of negative weights. On Fri, Dec 17, 2010 at 11:04 AM, Chris Schilling <[email protected]>wrote: > Hey Ted, > > Any word on this? Is there something I can do to help. I am just not real > sure on what side of the problem I am on: dissector code or learning > algorithm. > > > On Dec 16, 2010, at 5:02 PM, Chris Schilling wrote: > > > Hey Ted, > > > > Okay. I have tested in trunk and 0.4. Pretty similar results. > > > > > > On Dec 16, 2010, at 4:43 PM, Ted Dunning wrote: > > > >> I think that the confusion is that many of these have negative weights. > >> Thus god !=> sci.space, but windows => comp.windows.x. > >> > >> Are you running from trunk or 0.4? > >> > >> On Thu, Dec 16, 2010 at 4:24 PM, Chris Schilling <[email protected]> > wrote: > >> > >>> First few results of dissect() > >>> body=god -0.1 sci.space 4.0 -0.1394994576714021 > 5.0 > >>> -0.10322063352194852 > >>> body=atheists -0.1 comp.windows.x 5.0 -0.07383748917466922 > 1.0 > >>> -0.037205929610919175 > >>> body=christian -0.1 talk.politics.mideast 2.0 > >>> -0.029106552130967654 4.0 -0.0033808015660384875 > >>> body=he 0.1 talk.politics.mideast 18.0 0.07845100216340763 > 5.0 > >>> -0.011218075788326903 > >>> body=martin -0.1 talk.politics.mideast 7.0 > >>> -0.019407188307985972 10.0 0.00782255718617942 > >>> body=say -0.1 comp.sys.ibm.pc.hardware 4.0 > >>> -0.0480512351042981 17.0 0.0037854045183534166 > >>> body=windows 0.1 comp.windows.x 18.0 -0.06722265016470273 > 5.0 > >>> -0.009627757932247396 > >>> body=file -0.1 sci.med 7.0 -0.05790809278204335 5.0 > >>> -0.050492324263356765 > >>> body=government 0.1 talk.religion.misc 3.0 > >>> -0.06076111927305433 2.0 -0.052663471587524276 > >>> body=sale -0.1 talk.religion.misc 15.0 > >>> -0.03535708180324768 12.0 -0.03532746353789419 > >>> body=atheism -0.1 misc.forsale 8.0 -0.05941771751639946 > 1.0 > >>> -0.0500729187538798 > >>> body=program -0.1 sci.med 16.0 -0.03820018259936702 7.0 > >>> -2.9675316187177843E-4 > >>> body=193 0.1 talk.politics.mideast 5.0 > 0.05061582599095028 > >>> 17.0 -0.032606809778589076 > >>> body=his -0.1 talk.politics.misc 12.0 > 0.05030942352260737 > >>> 5.0 -0.04490996261214399 > >>> > >>> I am not adding any leaks (leakType = 0). > >>> > >>> Any ideas here? > >>> > > > >
