[
https://issues.apache.org/jira/browse/SPARK-5056?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14272277#comment-14272277
]
Xiangrui Meng commented on SPARK-5056:
--------------------------------------
[~tmilinovic] We had some discussion in SPARK-4510 about the complexity of
k-medoids possible solutions. Does the proposed algorithm have better
complexity?
> Implementing Clara k-medoids clustering algorithm for large datasets
> --------------------------------------------------------------------
>
> Key: SPARK-5056
> URL: https://issues.apache.org/jira/browse/SPARK-5056
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: Tomislav Milinovic
> Priority: Minor
> Labels: features
>
> There is a specific k-medoids clustering algorithm for large datasets. The
> algorithm is called Clara in R, and is fully described in chapter 3 of
> Finding Groups in Data: An Introduction to Cluster Analysis. by Kaufman, L
> and Rousseeuw, PJ (1990).
> The algorithm considers sub-datasets of fixed size (sampsize) such that the
> time and storage requirements become linear in n rather than quadratic. Each
> sub-dataset is partitioned into k clusters using the same algorithm as in
> Partinioning around Medoids (PAM).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
