You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Pat Ferrel (JIRA)" <ji...@apache.org> on 2012/05/24 01:15:40 UTC
[jira] [Created] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Pat Ferrel created MAHOUT-1020:
----------------------------------
Summary: The Cluster Evaluator is returning bad results
Key: MAHOUT-1020
URL: https://issues.apache.org/jira/browse/MAHOUT-1020
Project: Mahout
Issue Type: Bug
Components: Clustering
Affects Versions: 0.6
Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
Reporter: Pat Ferrel
Fix For: 0.7
Conversation with between Pat Ferrel and Jeff Eastman on the user list
Hi Pat,
I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
>From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
* Return if the cluster is valid. Valid clusters must have more than 2 representative points,
* and at least one of them must be different than the cluster center. This is because the
* representative points extraction will duplicate the cluster center if it is empty.
Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
Thanks,
Jeff
On 5/17/12 3:53 PM, Pat Ferrel wrote:
> I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
>
> When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13287549#comment-13287549 ]
Hudson commented on MAHOUT-1020:
--------------------------------
Integrated in Mahout-Quality #1510 (See [https://builds.apache.org/job/Mahout-Quality/1510/])
MAHOUT-1020: fixed path names for testKmeans and testFuzzyKmeans that were causing representative points calculation to fail. Evaluator results now look more reasonable. (Revision 1345269)
Result = FAILURE
jeastman : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1345269
Files :
* /mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterEvaluator.java
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Pat Ferrel (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13411935#comment-13411935 ]
Pat Ferrel commented on MAHOUT-1020:
------------------------------------
https://issues.apache.org/jira/browse/MAHOUT-1045
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6, 0.8
> Environment: Various environments and data sets. Mahout 0.6, 0.7, 0.8 trunk.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.8
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Jeff Eastman (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13287461#comment-13287461 ]
Jeff Eastman commented on MAHOUT-1020:
--------------------------------------
It looks like new kmeansOutput and fuzzyKMeansOutput paths were introduced for the clustering output and the representative points computation was not updated so none were produced. I've fixed those two tests and now the CDbw results look more respectable. Committing these changes now.
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13287466#comment-13287466 ]
Hudson commented on MAHOUT-1020:
--------------------------------
Integrated in Mahout-Quality #1509 (See [https://builds.apache.org/job/Mahout-Quality/1509/])
MAHOUT-1020: fixed path names for testKmeans and testFuzzyKmeans that were causing representative points calculation to fail. CDbw results now look more reasonable. (Revision 1345214)
Result = FAILURE
jeastman : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1345214
Files :
* /mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/cdbw/TestCDbwEvaluator.java
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Grant Ingersoll (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Grant Ingersoll resolved MAHOUT-1020.
-------------------------------------
Resolution: Fixed
I think the build failure is unrelated to this issue. So, closing this issue.
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Assigned] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Jeff Eastman (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jeff Eastman reassigned MAHOUT-1020:
------------------------------------
Assignee: Jeff Eastman
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Pat Ferrel (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13409696#comment-13409696 ]
Pat Ferrel commented on MAHOUT-1020:
------------------------------------
This issue is still a problem even though the build succeeds. Please reopen it.
I'll associate some data later but a short description is that with most real world data the Intra-cluster density from ClusterEvaluator is almost always NaN. The CDbw inter-cluster density is almost always 0. I have also seen several cases where CDbw fails to return any results but have not tracked down why yet.
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6, 0.8
> Environment: Various environments and data sets. Mahout 0.6, 0.7, 0.8 trunk.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.8
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Sean Owen (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13411897#comment-13411897 ]
Sean Owen commented on MAHOUT-1020:
-----------------------------------
(Since it's "Closed" we can't resurrect it in JIRA. Just start a new issue.)
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6, 0.8
> Environment: Various environments and data sets. Mahout 0.6, 0.7, 0.8 trunk.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.8
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13287765#comment-13287765 ]
Hudson commented on MAHOUT-1020:
--------------------------------
Integrated in Mahout-Quality #1513 (See [https://builds.apache.org/job/Mahout-Quality/1513/])
MAHOUT-1020: Refactored printRepresentativePoints utility to RepresentativePointsDriver as a static so it can be used more easily outside of the tests. All tests run. (Revision 1345375)
Result = FAILURE
jeastman : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1345375
Files :
* /mahout/trunk/integration/src/main/java/org/apache/mahout/clustering/evaluation/RepresentativePointsDriver.java
* /mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterEvaluator.java
* /mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/cdbw/TestCDbwEvaluator.java
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6
> Environment: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.7
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAHOUT-1020) The Cluster Evaluator is returning
bad results
Posted by "Pat Ferrel (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAHOUT-1020?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Pat Ferrel updated MAHOUT-1020:
-------------------------------
Fix Version/s: (was: 0.7)
0.8
Environment: Various environments and data sets. Mahout 0.6, 0.7, 0.8 trunk. (was: Various environments and data sets. Mahout 0.6, 0.7 trunk not tested.)
Affects Version/s: 0.8
> The Cluster Evaluator is returning bad results
> ----------------------------------------------
>
> Key: MAHOUT-1020
> URL: https://issues.apache.org/jira/browse/MAHOUT-1020
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.6, 0.8
> Environment: Various environments and data sets. Mahout 0.6, 0.7, 0.8 trunk.
> Reporter: Pat Ferrel
> Assignee: Jeff Eastman
> Fix For: 0.8
>
>
> Conversation with between Pat Ferrel and Jeff Eastman on the user list
> Hi Pat,
> I don't have a good answer here. Evidently, something in CDbw has become broken and you are the first to notice. When I run TestCDbwEvaluator, the values for k-means and fuzzy-k are clearly incorrect. The values for Canopy, MeanShift and Dirichlet are not so obviously incorrect but I remain suspicious. Something must have become broken in the recent clustering refactoring.
> From the method CDbwEvaluator.invalidCluster comment (used to enable pruning):
> * Return if the cluster is valid. Valid clusters must have more than 2 representative points,
> * and at least one of them must be different than the cluster center. This is because the
> * representative points extraction will duplicate the cluster center if it is empty.
> Oddly enough, inspection of the test log indicates that only k-means and fuzzy-k are not pruning clusters. Clearly some more investigation is needed. I will take a look at it tomorrow. In the mean time if you develop any additional insight please do share it with us.
> Thanks,
> Jeff
> On 5/17/12 3:53 PM, Pat Ferrel wrote:
> > I built a tool that iterates through a list of values for k on the same data and spits out the CDbw and ClusterEvaluator results each time.
> >
> > When the evaluator or CDbw prunes a cluster, how do I interpret that? They seem to throw out the same clusters on a given run. Also CDbw always returns an inter-cluster density of 0?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira