AN 300 Exam 2 Questions And Correct
cluster canalysis c- cCORRECT cANS✔✔a cdescriptive canalytics ctechnique cused cto cdiscover cnatur
c(answers c"what chas chappened?" cquestions)
what care cthe cproblem ccharacteristics cof ccluster canalysis? c- cCORRECT cANS✔✔1. chave cinform
cthe cobjects c(ex: ccustomers)
2. cno cprior cknowledge cof chow cthe cobjects care crelated cto ceach cother c(ex: cpurchasing cbehavio
3. cthe cobjective cis cto corganize cobjects cinto cgroups c(ex: cex: cmarket csegment)
what care cthe ctwo csimilarity cmeasures cand cwhat cdo cthey cmeasure? c- cCORRECT cANS✔✔1. cE
cnumerical cdata c(ex: cheight c& cweight)
2. cMatching cCoefficient cfor ccategorical cdata c(ex: clevels cof cincome) c
- cboth cgauge cwhether ca cgroup cof cobjects care csimilar cor cdissimilar cto cone canother
Euclidean cDistance c- cCORRECT cANS✔✔- cfor cnumerical cdata c
- cthe cdistance cbetween ctwo cobjects cis cthe clength cof ca cstraight clike cbetween cthem c
- cstandardize cthe cnumerical cdata cto cmake cit cunit-free cbefore ccalculating cthe cdistance cmeasure
Matching cCoefficient c- cCORRECT cANS✔✔- cfor ccategorical cdata c
- cnumber cof ccolumns cwith cmatching ccategorical cvalues/total cnumber cof ccolumns cof ccategorica
z-score c- cCORRECT cANS✔✔(raw cvalue c- cmean)/standard cdeviation
- cused cto cstandardize cdata cfor cEuclidean cDistance
what care csome cbusiness capplication cof ccluster canalysis? c- cCORRECT cANS✔✔1. cmarketing: cd
chomogeneous cgroups cfor ctarget cmarketing c
2. cfinance: cdivide cclients cinto chomogeneous cgroups cfor cpersonalized cfinance cadvice c
3. coperations: cidentify coutliers cfor cquality ccontrol
which cof cthe cfollowing cis ctrue cabout ccluster canalysis? c(check call cthat capply) c- cCORRECT cANS
chas chappened cquestions c
- cit cis cused cto cdiscover cnatural cgroupings cof cobjects c
- cit cis ca cdescriptive canalytics ctechnique
which cof cthe cfollowing cis ca ccharacteristic cof ca ccluster canalysis cproblem? c(check call cthat capply
cabout chow cto corganize cobjects cinto cgroups c
- cthe cdata cthat cdescribes cthe cobject cmust cbe cgiven c
- cits cobjective cis cto cmaximize csimilarities cof cobjects cwithin cgroups
you chave cdata con cthe cweight cand cheight cof cpatients. cwhich csimilarity cmeasure cshould cbe cuse
cgroup cof cpatients cis cto cone canother? c- cCORRECT cANS✔✔Euclidean cdistance
, Hierarchical cClustering c- cCORRECT cANS✔✔- cuseful cfor csmall cdata csets c(less cthan c500 crows
- csupports cnumeric cand ccategorical/binary cdata
- csensitive cto coutliers c
- cexperiment cwith cdifferent cmethods cto ccalculate cthe cdistance cbetween cclusters
K-Means cClustering c- cCORRECT cANS✔✔- cuseful cfor clarge cdata csets c(over c500 crows cof cdata
- csupports conly cnumeric cdata c
- cless csensitive cto coutliers c
- cexperiment cwith cdifferent cnumbers cof cclusters
how cdo cwe cestimate cthe cnumber cof cclusters? c- cCORRECT cANS✔✔cubic cclustering ccriterion c(
cubic cclustering ccriterion c(CCC) c- cCORRECT cANS✔✔- ca cmetric crelated cto cR2, cthe cproportion
caccounted cfor cby cthe cclusters c
- ca cCCC c> c2 cindicates cgood ccluster c
- ca cCCC cbetween c0-2 cindicates cpossible cclusters c
- clarge cnegative cvalues cof cCCC c(ex: c-30) cmay cbe cdue cto coutliers
how cdoes cthe chierarchical cclustering cprocess cwork? c- cCORRECT cANS✔✔- cstarts cwith cplacing
ccluster c
- citeratively ccombines ctwo cmost csimilar cclusters cinto cone c
- cstops cwhen call cobjects care cin cone ccluster
what care cthe cdifferent cdistance cmeasures cbetween cclusters? c- cCORRECT cANS✔✔- csingle clin
- ccomplete clinkage c(farthest cneighbor) c
- caverage clinkage c
- ccentroid clinkage c
- cWard's cminimum cvariance c(default cmeasure cin cJMP cPro)
single clinkage c- cCORRECT cANS✔✔- cnearest cneighbor c
- cthe cminimum cdistance cbetween ca cpair cof cobservations cthat cdo cnot cbelong cto cthe csame cclus
- ctends cto cproduce celongated cand cirregularly cshaped cclusters
complete clinkage c- cCORRECT cANS✔✔- cfarthest cneighbor c
- cthe cmaximum cdistance cbetween ca cpair cof cobservations cthat cdo cno cbelong cto cthe csame cclus
- ctends cto cproduce cclusters cwith cequal cdiameters cand csensitive cto coutliers
average clinkage c- cCORRECT cANS✔✔- cthe caverage cdistance cbetween call cpairs cof cobservation
csame ccluster c
- ctends cto cproduce cclusters cwith cthe csame cvariance
centroid clinkage c- cCORRECT cANS✔✔- cthe csquared cEuclidean cdistance cbetween cthe ccluster cm
- cleast csensitive cto coutliers
Ward's cminimum cvariance c- cCORRECT cANS✔✔- cdefault cmeasure cin cJMP cPro c
- cthe cminimum cerror csum cof csquare cwithin cclusters c
- ctends cto cproduce cclusters cwith cthe csame cnumber cof cobservations
what care cways cto cvisualize cclusters? c- cCORRECT cANS✔✔- cdendrogram c(tree-like)
- cdistance cgraph c(2-D)
- cconstellation cplot c(vein-like)
how cis ca cdendrogram cstructured? c- cCORRECT cANS✔✔1. cend cpoint c= clabel cof cthe cobject c
2. cvertical cline c= ccluster cgroupings c
3. chorizontal cline c= crelative cdistance cbetween cclusters c
cluster canalysis c- cCORRECT cANS✔✔a cdescriptive canalytics ctechnique cused cto cdiscover cnatur
c(answers c"what chas chappened?" cquestions)
what care cthe cproblem ccharacteristics cof ccluster canalysis? c- cCORRECT cANS✔✔1. chave cinform
cthe cobjects c(ex: ccustomers)
2. cno cprior cknowledge cof chow cthe cobjects care crelated cto ceach cother c(ex: cpurchasing cbehavio
3. cthe cobjective cis cto corganize cobjects cinto cgroups c(ex: cex: cmarket csegment)
what care cthe ctwo csimilarity cmeasures cand cwhat cdo cthey cmeasure? c- cCORRECT cANS✔✔1. cE
cnumerical cdata c(ex: cheight c& cweight)
2. cMatching cCoefficient cfor ccategorical cdata c(ex: clevels cof cincome) c
- cboth cgauge cwhether ca cgroup cof cobjects care csimilar cor cdissimilar cto cone canother
Euclidean cDistance c- cCORRECT cANS✔✔- cfor cnumerical cdata c
- cthe cdistance cbetween ctwo cobjects cis cthe clength cof ca cstraight clike cbetween cthem c
- cstandardize cthe cnumerical cdata cto cmake cit cunit-free cbefore ccalculating cthe cdistance cmeasure
Matching cCoefficient c- cCORRECT cANS✔✔- cfor ccategorical cdata c
- cnumber cof ccolumns cwith cmatching ccategorical cvalues/total cnumber cof ccolumns cof ccategorica
z-score c- cCORRECT cANS✔✔(raw cvalue c- cmean)/standard cdeviation
- cused cto cstandardize cdata cfor cEuclidean cDistance
what care csome cbusiness capplication cof ccluster canalysis? c- cCORRECT cANS✔✔1. cmarketing: cd
chomogeneous cgroups cfor ctarget cmarketing c
2. cfinance: cdivide cclients cinto chomogeneous cgroups cfor cpersonalized cfinance cadvice c
3. coperations: cidentify coutliers cfor cquality ccontrol
which cof cthe cfollowing cis ctrue cabout ccluster canalysis? c(check call cthat capply) c- cCORRECT cANS
chas chappened cquestions c
- cit cis cused cto cdiscover cnatural cgroupings cof cobjects c
- cit cis ca cdescriptive canalytics ctechnique
which cof cthe cfollowing cis ca ccharacteristic cof ca ccluster canalysis cproblem? c(check call cthat capply
cabout chow cto corganize cobjects cinto cgroups c
- cthe cdata cthat cdescribes cthe cobject cmust cbe cgiven c
- cits cobjective cis cto cmaximize csimilarities cof cobjects cwithin cgroups
you chave cdata con cthe cweight cand cheight cof cpatients. cwhich csimilarity cmeasure cshould cbe cuse
cgroup cof cpatients cis cto cone canother? c- cCORRECT cANS✔✔Euclidean cdistance
, Hierarchical cClustering c- cCORRECT cANS✔✔- cuseful cfor csmall cdata csets c(less cthan c500 crows
- csupports cnumeric cand ccategorical/binary cdata
- csensitive cto coutliers c
- cexperiment cwith cdifferent cmethods cto ccalculate cthe cdistance cbetween cclusters
K-Means cClustering c- cCORRECT cANS✔✔- cuseful cfor clarge cdata csets c(over c500 crows cof cdata
- csupports conly cnumeric cdata c
- cless csensitive cto coutliers c
- cexperiment cwith cdifferent cnumbers cof cclusters
how cdo cwe cestimate cthe cnumber cof cclusters? c- cCORRECT cANS✔✔cubic cclustering ccriterion c(
cubic cclustering ccriterion c(CCC) c- cCORRECT cANS✔✔- ca cmetric crelated cto cR2, cthe cproportion
caccounted cfor cby cthe cclusters c
- ca cCCC c> c2 cindicates cgood ccluster c
- ca cCCC cbetween c0-2 cindicates cpossible cclusters c
- clarge cnegative cvalues cof cCCC c(ex: c-30) cmay cbe cdue cto coutliers
how cdoes cthe chierarchical cclustering cprocess cwork? c- cCORRECT cANS✔✔- cstarts cwith cplacing
ccluster c
- citeratively ccombines ctwo cmost csimilar cclusters cinto cone c
- cstops cwhen call cobjects care cin cone ccluster
what care cthe cdifferent cdistance cmeasures cbetween cclusters? c- cCORRECT cANS✔✔- csingle clin
- ccomplete clinkage c(farthest cneighbor) c
- caverage clinkage c
- ccentroid clinkage c
- cWard's cminimum cvariance c(default cmeasure cin cJMP cPro)
single clinkage c- cCORRECT cANS✔✔- cnearest cneighbor c
- cthe cminimum cdistance cbetween ca cpair cof cobservations cthat cdo cnot cbelong cto cthe csame cclus
- ctends cto cproduce celongated cand cirregularly cshaped cclusters
complete clinkage c- cCORRECT cANS✔✔- cfarthest cneighbor c
- cthe cmaximum cdistance cbetween ca cpair cof cobservations cthat cdo cno cbelong cto cthe csame cclus
- ctends cto cproduce cclusters cwith cequal cdiameters cand csensitive cto coutliers
average clinkage c- cCORRECT cANS✔✔- cthe caverage cdistance cbetween call cpairs cof cobservation
csame ccluster c
- ctends cto cproduce cclusters cwith cthe csame cvariance
centroid clinkage c- cCORRECT cANS✔✔- cthe csquared cEuclidean cdistance cbetween cthe ccluster cm
- cleast csensitive cto coutliers
Ward's cminimum cvariance c- cCORRECT cANS✔✔- cdefault cmeasure cin cJMP cPro c
- cthe cminimum cerror csum cof csquare cwithin cclusters c
- ctends cto cproduce cclusters cwith cthe csame cnumber cof cobservations
what care cways cto cvisualize cclusters? c- cCORRECT cANS✔✔- cdendrogram c(tree-like)
- cdistance cgraph c(2-D)
- cconstellation cplot c(vein-like)
how cis ca cdendrogram cstructured? c- cCORRECT cANS✔✔1. cend cpoint c= clabel cof cthe cobject c
2. cvertical cline c= ccluster cgroupings c
3. chorizontal cline c= crelative cdistance cbetween cclusters c
