[ 
https://issues.apache.org/jira/browse/MAHOUT-1045?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13414675#comment-13414675
 ] 

Jeff Eastman commented on MAHOUT-1045:
--------------------------------------

I tried changing the initialization of the max loop variable from 0 to 
Double.MIN_VALUE (see below) and now it produces a legal value (0.656) for 
average Intra-Cluster Density. Is this useful? Does it look like a reasonable 
change?

{code}
  public double intraClusterDensity() {
    pruneInvalidClusters();
    double avgDensity = 0;
    for (Cluster cluster : clusters) {
      int count = 0;
      double max = Double.MIN_VALUE;
      double min = Double.MAX_VALUE;
      double sum = 0;
      List<VectorWritable> repPoints = 
representativePoints.get(cluster.getId());
      for (int i = 0; i < repPoints.size(); i++) {
        for (int j = i + 1; j < repPoints.size(); j++) {
          double d = measure.distance(repPoints.get(i).get(), 
repPoints.get(j).get());
          min = Math.min(d, min);
          max = Math.max(d, max);
          sum += d;
          count++;
        }
      }
      double denom = max - min;
      double density = (sum / count - min) / denom;
      avgDensity += density;
      log.info("Intra-Cluster Density[{}] = {}", cluster.getId(), density);
    }
    avgDensity = clusters.isEmpty() ? 0 : avgDensity / clusters.size();
    log.info("Intra-Cluster Density = {}", avgDensity);
    return avgDensity;   
  }
{code}
                
> Cluster evaluators returning bad results
> ----------------------------------------
>
>                 Key: MAHOUT-1045
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1045
>             Project: Mahout
>          Issue Type: Bug
>          Components: Clustering
>    Affects Versions: 0.6, 0.7, 0.8
>         Environment: Several environments and data sets
>            Reporter: Pat Ferrel
>             Fix For: 0.8
>
>
> With real world crawl 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.
> I have sent a link to an 8G data set that reproduces these errors to Jeff 
> Eastman.

--
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

        

Reply via email to