Glad it's working better now. The results are about what I would expect. Some
empty clusters indicates k was set high enough to capture the important models,
given the alpha0 setting. The large number of documents in DC-0 suggests
adjusting a0, while increasing k, could find more subtle structure within your
data.




On October 28, 2011 at 1:29 AM edward choi <[email protected]> wrote:

> Okay, I have tested with Reuters set and the result was much better than
> testing with my news documents.
>
> I downloaded Reuters set, made it into sequence file. Then turned it into
> sparse vector with following arguments:
> --minDF 2 --maxDFPercent 50 --weight TFIDF --norm 2 -ng 2 -nv
> Then I did DPC with the same arguments you told me.
>
> The total number of documents was 21578.
> DC-0 had 11187 documents.
> Seven clusters had zero docs.
> Rest of the clusters had from 1 to 1189 docs.
>
> Very interesting thing is, DC-16,18, 19 have the exact same negative points
> as before when I did DPC with my own document set.
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> DC-16 total= 0 model= DMC:16{n=0 c=[0:-0.411, 0.003:-0.061, 0.01:1.685,
> 0.02:-0.560, 0.025:-0.147, 0.03:-0.675, 0.04:-0.234, 0.05:-0.430,
> 0.06:0.451, 0.07:0.186, 0.073:-0.799, 0.077:0.724, 0.1:0.731, 0.10:2.274,
> 0.11:-0.739, 0.12:0.660, 0.127:1.546, 0.13:0.907, 0.139:0.839, 0.14:-0.060,
> 0.15:0.006, 0.16:0.294, 0.163:-0.458, 0.17:0.057, 0.18:0.173, 0.185:0.938,
> 0.19:-1.340, 0.194:-0.597, 0.2:0.311, 0.20:-0.318, 0.206:-0.053,
> 0.21:-0.198, 0.2125:-1.851, 0.22:-0.604,................
>     Top Terms:
> jersey based                            =>   5.055564881106928
> withdrew offer                          =>   4.160793145890344
> although said                           =>  4.1069074456260966
> confirmed iraqi                         =>   4.016748531705415
> force administration                    =>   3.995899196620034
> 24.6                                    =>  3.9719147317695596
> due mostly                              =>  3.9125799367453267
> unit british                            =>  3.9048586110602286
> trade source                            =>   3.892495010521945
> stevens                                 =>  3.7816279439782554
> DC-18 total= 0 model= DMC:18{n=0 c=[0:-0.411, 0.003:-0.061, 0.01:1.685,
> 0.02:-0.560, 0.025:-0.147, 0.03:-0.675, 0.04:-0.234, 0.05:-0.430,
> 0.06:0.451, 0.07:0.186, 0.073:-0.799, 0.077:0.724, 0.1:0.731, 0.10:2.274,
> 0.11:-0.739, 0.12:0.660, 0.127:1.546, 0.13:0.907, 0.139:0.839, 0.14:-0.060,
> 0.15:0.006, 0.16:0.294, 0.163:-0.458, 0.17:0.057, 0.18:0.173, 0.185:0.938,
> 0.19:-1.340, 0.194:-0.597, 0.2:0.311, 0.20:-0.318, 0.206:-0.053,
> 0.21:-0.198, 0.2125:-1.851, 0.22:-0.604,..............
>     Top Terms:
> jersey based                            =>   5.055564881106928
> withdrew offer                          =>   4.160793145890344
> although said                           =>  4.1069074456260966
> confirmed iraqi                         =>   4.016748531705415
> force administration                    =>   3.995899196620034
> 24.6                                    =>  3.9719147317695596
> due mostly                              =>  3.9125799367453267
> unit british                            =>  3.9048586110602286
> trade source                            =>   3.892495010521945
> stevens                                 =>  3.7816279439782554
> DC-19 total= 0 model= DMC:19{n=0 c=[0:-0.411, 0.003:-0.061, 0.01:1.685,
> 0.02:-0.560, 0.025:-0.147, 0.03:-0.675, 0.04:-0.234, 0.05:-0.430,
> 0.06:0.451, 0.07:0.186, 0.073:-0.799, 0.077:0.724, 0.1:0.731, 0.10:2.274,
> 0.11:-0.739, 0.12:0.660, 0.127:1.546, 0.13:0.907, 0.139:0.839, 0.14:-0.060,
> 0.15:0.006, 0.16:0.294, 0.163:-0.458, 0.17:0.057, 0.18:0.173, 0.185:0.938,
> 0.19:-1.340, 0.194:-0.597, 0.2:0.311, 0.20:-0.318, 0.206:-0.053,
> 0.21:-0.198, 0.2125:-1.851, 0.22:-0.604,...........
>     Top Terms:
> jersey based                            =>   5.055564881106928
> withdrew offer                          =>   4.160793145890344
> although said                           =>  4.1069074456260966
> confirmed iraqi                         =>   4.016748531705415
> force administration                    =>   3.995899196620034
> 24.6                                    =>  3.9719147317695596
> due mostly                              =>  3.9125799367453267
> unit british                            =>  3.9048586110602286
> trade source                            =>   3.892495010521945
> stevens                                 =>  3.7816279439782554
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> So I'm guessing there is some kind of algorithmic problem since the test
> sets were different but the same DC-16,18,19 have the same values?
>
> Regards,
> Ed
>
> 2011/10/28 edward choi <[email protected]>
>
> >
> > I downloaded the most recent version of Mahout from apache SVN.
> > Using the new arguments, I have tested DPC on my own news documents. (not
> > reuters set)
> >
> > Turns out, it really had great improvements. First of all, documents are
> > somewhat distributed across 20 clusters.
> > The total number of documents were 5896.
> > DC-0 had 1014 documents. DC-1 had 4305 documents.
> > Nine clusters had zero documents. Rest of the clusters had from 1 to 214
> > documents each.
> >
> > The quality of the clusters weren't so pretty but I guess that has got to
> > do with the crude preprocessing step. (raw news documents have links, ads,
> > reader comments, etc. etc. etc)
> > I will know better when I test with build-reuters.sh
> >
> > One more thing. Unfortunately there are still some negative values in the
> > cluster points.
> >
> > -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > DC-16 total= 0 model= DMC:16{n=0 c=[0:-1.093, 0.07:-0.891, 0.08:1.327,
> > 0.1:0.504, 0.18:-0.705, 0.2:0.318, 0.25:1.824, 0.3:0.273, 0.32:-0.792,
> > 0.4:0.390, 0.41:-1.314, 0.5:0.727, 0.7:0.734, 0.70:-0.973,
> >     Top Terms:
> >         kodak camera                            =>  4.5009259007672835
> >         player july                             =>   4.216287519075373
> >         figure mix                              =>   4.139826527167421
> >         department defense                      =>   4.009974576583582
> >         remark wednesday                        =>  3.9945681051149564
> >         counsel infection                       =>   3.886000915158471
> >         jefferson county                        =>  3.8442975919513667
> >         jersey say                              =>  3.7821696224124786
> >         tell couple                             =>  3.7644857721992415
> >         3.5 million                             =>   3.743525174300145
> > DC-18 total= 0 model= DMC:18{n=0 c=[0:-1.093, 0.07:-0.891, 0.08:1.327,
> > 0.1:0.504, 0.18:-0.705, 0.2:0.318, 0.25:1.824, 0.3:0.273, 0.32:-0.792,
> > 0.4:0.390, 0.41:-1.314, 0.5:0.727, 0.7:0.734, 0.70:-0.973,
> >     Top Terms:
> >         kodak camera                            =>  4.5009259007672835
> >         player july                             =>   4.216287519075373
> >         figure mix                              =>   4.139826527167421
> >         department defense                      =>   4.009974576583582
> >         remark wednesday                        =>  3.9945681051149564
> >         counsel infection                       =>   3.886000915158471
> >         jefferson county                        =>  3.8442975919513667
> >         jersey say                              =>  3.7821696224124786
> >         tell couple                             =>  3.7644857721992415
> >         3.5 million                             =>   3.743525174300145
> > DC-19 total= 0 model= DMC:19{n=0 c=[0:-1.093, 0.07:-0.891, 0.08:1.327,
> > 0.1:0.504, 0.18:-0.705, 0.2:0.318, 0.25:1.824, 0.3:0.273, 0.32:-0.792,
> > 0.4:0.390, 0.41:-1.314, 0.5:0.727, 0.7:0.734, 0.70:-0.973,
> >     Top Terms:
> >         kodak camera                            =>  4.5009259007672835
> >         player july                             =>   4.216287519075373
> >         figure mix                              =>   4.139826527167421
> >         department defense                      =>   4.009974576583582
> >         remark wednesday                        =>  3.9945681051149564
> >         counsel infection                       =>   3.886000915158471
> >         jefferson county                        =>  3.8442975919513667
> >         jersey say                              =>  3.7821696224124786
> >         tell couple                             =>  3.7644857721992415
> >         3.5 million                             =>   3.743525174300145
> >
> > -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > Among nine clusters which have zero members, above three have negative
> > values.
> > Interestingly, all three of them have the exact same values and top terms.
> > I wonder what this means.
> >
> > Anyway I'll post another thread when I have played around with Reuters set.
> >
> > Ed
> >
> > ps. The runtime has indeed reduced significantly!!! Possibly 100 times
> > faster as you said. Loved it!!
> >
> > 2011/10/20 Jeff Eastman <[email protected]>
> >
> >> R1186452 commits two small changes that seem to do much better with
> >> Reuters than before:
> >> - fixed DistanceMeasureClusterDistribution to generate Gaussian element
> >> values in the prior clusters. Zero values in previous implementation don't
> >> work with CosineDistanceMeasure.
> >> - changed Dirichlet arguments to use DMCD and CosineDM in build-reuters.sh
> >> - switched -mp to DenseVector since all the prior center elements are
> >> Gaussian and generally non-zero
> >> - increased -a0 to 2
> >>
> >> Build-reuters now does a much better job with the wide topic vectors using
> >> the DMCD/CosineDM. And it runs maybe 100x faster too. Here are the new
> >> arguments:
> >>
> >>  $MAHOUT dirichlet \
> >>    -i ${WORK_DIR}/reuters-out-seqdir-sparse-dirichlet/tfidf-vectors \
> >>    -o ${WORK_DIR}/reuters-dirichlet -k 20 -ow -x 10 -a0 2 \
> >>    -md
> >> org.apache.mahout.clustering.dirichlet.models.DistanceMeasureClusterDistribution
> >> \
> >>    -mp org.apache.mahout.math.DenseVector \
> >>    -dm org.apache.mahout.common.distance.CosineDistanceMeasure
> >>
> >>
> >> -----Original Message-----
> >> From: Jeff Eastman [mailto:[email protected]]
> >> Sent: Wednesday, October 19, 2011 9:53 AM
> >> To: [email protected]
> >> Subject: RE: Dirichlet Process Clustering not working
> >>
> >> The pdf() implementation in GaussianCluster is pretty lame. It is
> >> computing a running product of the element pdfs which, for wide input
> >> vectors (Reuters is 41,807), always underflows and returns 0. Here's the
> >> code:
> >>
> >>  public double pdf(VectorWritable vw) {
> >>    Vector x = vw.get();
> >>    // return the product of the component pdfs
> >>    // TODO: is this reasonable? correct? It seems to work in some cases.
> >>    double pdf = 1;
> >>    for (int i = 0; i < x.size(); i++) {
> >>      // small prior on stdDev to avoid numeric instability when stdDev==0
> >>      pdf *= UncommonDistributions.dNorm(x.getQuick(i),
> >>          getCenter().getQuick(i), getRadius().getQuick(i) + 0.000001);
> >>    }
> >>    return pdf;
> >>  }
> >>
> >> -----Original Message-----
> >> From: Jeff Eastman [mailto:[email protected]]
> >> Sent: Wednesday, October 19, 2011 9:04 AM
> >> To: [email protected]
> >> Subject: RE: Dirichlet Process Clustering not working
> >>
> >> I agree something is amiss here, but it could be the model is just not
> >> suitable for this problem. Running with the Reuters dataset, I see all the
> >> points being assigned to C-0 in the very first iteration as you do. I think
> >> the problem is with the pdf() calculations in the mapper for very wide
> >> vectors such as we are using. For smaller dimension vectors, DPC appears to
> >> be working great.
> >>
> >> I'm going to commit the build-reuters.sh enhancements I've added for
> >> FuzzyK and DPC so we can both use the same platform. I will report more
> >> progress as I dig in deeper today...
> >>
> >> -----Original Message-----
> >> From: edward choi [mailto:[email protected]]
> >> Sent: Wednesday, October 19, 2011 8:11 AM
> >> To: [email protected]
> >> Subject: Re: Dirichlet Process Clustering not working
> >>
> >> Okay, I've just tried DPC with reuters document set.
> >> I let the 'build-reuters.sh' create the sequence files and vectors. (From
> >> the looks of its dictionary generated by mahout, the number of features
> >> seemed to be less than 100,000)
> >> Then I used them to do DPC. (15 clusters, 10 iteration, 1.0 alpha,
> >> clustering true, no addtional options)
> >> Below is the result of the clusterdump of clusters-10
> >>
> >> ----------------------------------------------------------------------------------------------------------------------------
> >> C-0: GC:0{n=15745 c=[0:0.026, 0.003:0.001, 0.01:0.004, 0.02:0.002,
> >> 0.05:0.004, 0.07:0.005, 0.07
> >>    Top Terms:
> >>        said                                    =>  1.6577128281476725
> >>        mln                                     =>  1.2455441154347937
> >>        dlrs                                    =>  1.1173752482257673
> >>        3                                       =>   1.042824193090437
> >>        pct                                     =>  1.0223684722334667
> >>        reuter                                  =>  0.9934255143959358
> >> C-1: GC:1{n=0 c=[0:-0.595, 0.003:0.228, 0.01:-0.401, 0.02:-0.711,
> >> 0.05:1.840, 0.07:0.136, 0.077:-0.739, 0.1:-0.177, 0.10:
> >>    Top Terms:....
> >> C-10: GC:10{n=0 c=[0:0.090, 0.003:-1.426, 0.01:-0.472, 0.02:0.672,
> >> 0.05:0.800, 0.07:0.691, 0.077:1.037, 0.1:0
> >>    Top Terms:....
> >> C-11: GC:11{n=0 c=[0:-0.835, 0.003:-1.748, 0.01:-1.030, 0.02:-1.760,
> >> 0.05:-0.343, 0.07:0.286, 0.077:1.179,
> >>    Top Terms:....
> >>
> >> ----------------------------------------------------------------------------------------------------------------------------
> >> I guess the same thing happened again. So the document set is not the
> >> problem. Something is definitely wrong with DPC.
> >> Interesting thing is that the first cluster point does not have a single
> >> negative value in it.
> >> Rest of the cluster points have a lot of negative values. So I guess this
> >> phenomenon has something to do with the first cluster hogging all the
> >> documents.
> >> Any comments on this result?
> >> (I haven't tried TestClusterDumper.testDirichlet2&3 yet. I'll post another
> >> thread when I am done with that).
> >>
> >> Regards,
> >> Ed
> >>
> >>
> >>
> >

Reply via email to