You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by Vijaypal Singh <s....@gmail.com> on 2013/02/18 06:24:27 UTC

Unable to understand LDA CBV output

Hi,

I executed the follwing step on my input data to run lda cvb.

I am using mahout 0.6 on llinux

./mahout seq2sparse --input /cvb/input/mail_clean/seq --output
/cvb/input/mail_clean/sparse -seq -wt tf -nr 1
-nv -a com.analyzer.Preprocessor -s 1

./mahout rowid --input /cvb/input/mail_clean/sparse/tf-vectors --output
/cvb/input/mail_clean/sparse-vectors-cvb

bin/hadoop fs -mv /cvb/input/mail_clean/sparse-vectors-cvb/docIndex
/cvb/input/mail_clean/sparse-vectors-index-cvb

./mahout cvb --input /cvb/input/mail_clean/sparse-vectors-cvb --output
/cvb/output/ftt-cvb -k 150
-ow -x 10 -dict /cvb/input/mail_clean/sparse/dictionary.file-0 -mt
/cvb/output/topic-model-cvb -dt /cvb/output/doc-topic-cvb

./mahout seqdumper -seq /cvb/input/mail_clean/sparse-vectors-cvb/matrix

./mahout ldatopics --input /cvb/output/topic-model-cvb/model-10 -d
/cvb/input/mail_clean/sparse/dictionary.file-0 -dt sequencefile -w 10

But I cannot make sens of the output generated by the last two steps.

Can anbody please help.

./mahout vectordump --seqFile
/cvb/output/topic-model-cvb/model-10/part-r-00000 --dictionary
/cvb/input/mail_clean/sparse/dictionary.file-0 --dictionaryType
sequencefile -n 10 --sortVectors true

{aaa:0.0022570866641009515,aac:0.0076465004730879495,aacc:0.009558705575006193,aacsb:0.004805255460414592,aad:0.007255394438432735,aadvisor:0.0023029941174043823,aag:0.006297946208626921,aai:0.003094778736901049,aam:0.00375279921096809,aamssm:0.0013482370771316983,aapl:0.008216861712751815,aaron:0.024000507569045478,aarp:0.002721215584105251,aba:0.0040347480431600635,abacu:0.007570044440629809,abad:0.0049168702169941495,abalx:0.003171097490885102,abandon:0.011141839881911687,abb:0.0031423831701993267,abbac:0.004302262499506318,abbett:0.005228171690004222,abbi:0.01676958339757493,abbot:0.02648426822312055,abbott:0.007910922271504386,abbrevi:0.010485621967884233,abc:0.0012916037917927385,abdukadir:0.0011776869315566114,abecu:0.00166690478412305,abel:4.91826935423341E-4,aberdeen:0.0064485603976139816,abernathi:0.006229867351660679,abf:0.010100576859206796,abfb:0.010026708399693867,abg:0.0077787535003316114,abicx:2.575071616931705E-4,abid:0.03547111807868174,abil:0.0030680752714131432,abilen:0.00579831466995624,abington:0.003042030994373309,abiquiu:0.00535356094271439,abl:0.008468224831453694,abndx:0.0021969752542052336,abner:0.008939743280886748,abo:6.684220420570368E-4,aboard:0.006239303620259144,abombeck:0.011236134727688458,abort:0.009498171173720793,abou:0.0049966979115371075,abraham:0.006240935779156503,abrahamson:0.008939338506347943,abramson:0.004674578187803808,abreast:0.005842924347685133,abreu:0.002667194167212146,abroad:0.012821041842075127,absenc:0.006046379046663534,absent:0.0020989858226379498,absente:0.004281989238231343,absolut:0.006300563556270954,absorb:0.0026602261802111658,abstract:0.005079275305504443,abt:0.0014395347695340855,abu:0.003230273747642934,abund:0.0031699364807168382,abus:7.100268602180975E-4,abut:0.0036248532928783686,abysm:0.0024333123222414934,aca:0.0017353779302128362,acad:0.014412715862681766,academ:0.0033939785016512286,academi:0.00913704409477667,acat:0.0024286034479600164,acc:0.006734735852738092,acccept:0.004607535214635505,acceler:0.0011249078590427942,accept:0.004838434578376805,acceso:0.0032501465782606427,access:0.014466429745069388,accid:0.003678010962633658,acciden:0.002894232588684375,accident:0.0019302096793321222,acclaim:0.006390169362383269,acclim:0.00220037225879197,acco:0.009107222567322073,accolad:0.005618595538131841,accommod:0.01560406612140553,accomp:0.007080021074839861,accompani:0.005340315644998393,accomplish:7.641799541535988E-4,accord:4.962053848220906E-4,accordingli:0.009146813317090921,accost:0.0038126311950500602,accou:0.010634699742351804,accoun:0.007112174246473042,account:0.01295624417628005,accounthold:0.0094824524
4,achiev:0.017534923057458643,achieveglob:0.002190134772520296,aci:0.008065237787711598,ackerman:0.007212081175493321,acknowledg:0.0035940914602369213,acl:0.0017651936863457012,acm:0.0035695424953583197,acopi:0.004082539596707424,acor:0.001141122797466815,acorn:6.197957279041173E-4,acquaint:0.0039886567545173645,acquir:0.0025230009863920055,acquisit:0.00443951641766913,acr:0.01014703560319759,acro:0.0053046101134567435,act:0.003118953835202118,actio:0.052245229770828024,action:0.013920806130158555,activ:0.0013307710882725426,activemq:0.0025262366996061355,activevo:0.00967814712687089,activist:0.006929836109946518,acton:0.00653139924148442,actual:0.0013722618645753048,actuari:0.00469718651021103,acupuncturist:0.0024119885334816025,acut:0.009677360503405874,acw:0.0030902817736559954,acworth:0.0018842234012178545,ada:0.005642902755499659,adam:0.002846882305565752,adamczyk:0.00211891539469639,adamson:0.02162081285213779,adan:0.0010269367883341538,adandrew:7.506284472368343E-4,adanhyperlink:0.008607165437613588,adapt:0.0026794986354330692,adaptor:0.020568175095557955,adcock:0.001729880835647783,add:0.0032750681178933157,addi:0.005633284632679811}



./mahout vectordump --seqFile /cvb/output/doc-topic-cvb/part-m-00000
--dictionary /cvb/input/mail_clean/sparse/dictionary.file-0
--dictionaryType sequencefile -n 10 --vectorSize 10 --sortVectors

{aaa:0.0022570866641009515,aac:0.0076465004730879495,aacc:0.009558705575006193,aacsb:0.004805255460414592,aad:0.007255394438432735,aadvisor:0.0023029941174043823,aag:0.006297946208626921,aai:0.003094778736901049,aam:0.00375279921096809,aamssm:0.0013482370771316983,aapl:0.008216861712751815,aaron:0.024000507569045478,aarp:0.002721215584105251,aba:0.0040347480431600635,abacu:0.007570044440629809,abad:0.0049168702169941495,abalx:0.003171097490885102,abandon:0.011141839881911687,abb:0.0031423831701993267,abbac:0.004302262499506318,abbett:0.005228171690004222,abbi:0.01676958339757493,abbot:0.02648426822312055,abbott:0.007910922271504386,abbrevi:0.010485621967884233,abc:0.0012916037917927385,abdukadir:0.0011776869315566114,abecu:0.00166690478412305,abel:4.91826935423341E-4,aberdeen:0.0064485603976139816,abernathi:0.006229867351660679,abf:0.010100576859206796,abfb:0.010026708399693867,abg:0.0077787535003316114,abicx:2.575071616931705E-4,abid:0.03547111807868174,abil:0.0030680752714131432,abilen:0.00579831466995624,abington:0.003042030994373309,abiquiu:0.00535356094271439,abl:0.008468224831453694,abndx:0.0021969752542052336,abner:0.008939743280886748,abo:6.684220420570368E-4,aboard:0.006239303620259144,abombeck:0.011236134727688458,abort:0.009498171173720793,abou:0.0049966979115371075,abraham:0.006240935779156503,abrahamson:0.008939338506347943,abramson:0.004674578187803808,abreast:0.005842924347685133,abreu:0.002667194167212146,abroad:0.012821041842075127,absenc:0.006046379046663534,absent:0.0020989858226379498,absente:0.004281989238231343,absolut:0.006300563556270954,absorb:0.0026602261802111658,abstract:0.005079275305504443,abt:0.0014395347695340855,abu:0.003230273747642934,abund:0.0031699364807168382,abus:7.100268602180975E-4,abut:0.0036248532928783686,abysm:0.0024333123222414934,aca:0.0017353779302128362,acad:0.014412715862681766,academ:0.0033939785016512286,academi:0.00913704409477667,acat:0.0024286034479600164,acc:0.006734735852738092,acccept:0.004607535214635505,acceler:0.0011249078590427942,accept:0.004838434578376805,acceso:0.0032501465782606427,access:0.014466429745069388,accid:0.003678010962633658,acciden:0.002894232588684375,accident:0.0019302096793321222,acclaim:0.006390169362383269,acclim:0.00220037225879197,acco:0.009107222567322073,accolad:0.005618595538131841,accommod:0.01560406612140553,accomp:0.007080021074839861,accompani:0.005340315644998393,accomplish:7.641799541535988E-4,accord:4.962053848220906E-4,accordingli:0.009146813317090921,accost:0.0038126311950500602,accou:0.010634699742351804,accoun:0.007112174246473042,account:0.01295624417628005,accounthold:0.009482452449486629,accountlink:0.0024611429485143564,accountsto:0.010617638274278453,accountwithout:0.0058404621805158855,accr:0.004942214601995807,accredit:0.034855545876910254,accru:0.00402849497254272,acct:0.01581620461785744,accumul:0.006770787845250496,accur:0.015722636919647006,accuraci:0.015580223684404986,accus:0.006866255550715339,acda:0.001138282150655383,acf:7.777883284080669E-4,ach:9.872453125907507E-4,achiev:0.017534923057458643,achieveglob:0.002190134772520296,aci:0.008065237787711598,ackerman:0.007212081175493321,acknowledg:0.0035940914602369213,acl:0.0017651936863457012,acm:0.0035695424953583197,acopi:0.004082539596707424,acor:0.001141122797466815,acorn:6.197957279041173E-4,acquaint:0.0039886567545173645,acquir:0.0025230009863920055,acquisit:0.00443951641766913,acr:0.01014703560319759,acro:0.0053046101134567435,act:0.003118953835202118,actio:0.052245229770828024,action:0.013920806130158555,activ:0.0013307710882725426,activemq:0.0025262366996061355,activevo:0.00967814712687089,activist:0.006929836109946518,acton:0.00653139924148442,actual:0.0013722618645753048,actuari:0.00469718651021103,acupuncturist:0.0024119885334816025,acut:0.009677360503405874,acw:0.0030902817736559954,acworth:0.0018842234012178545,ada:0.005642902755499659,adam:0.002846882305565752,adamczyk:0.00211891539469639,adamson:0.02162081285213779,adan:0.0010269367883341538,adandrew:7.506284472368343E-4,adanhyperlink:0.008607165437613588,adapt:0.0026794986354330692,adaptor:0.020568175095557955,adcock:0.001729880835647783,add:0.0032750681178933157,addi:0.005633284632679811}





-- 
Thanks and Regards

Vijaypal Singh