[
https://issues.apache.org/jira/browse/SPARK-15880?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15830137#comment-15830137
]
Takeshi Yamamuro commented on SPARK-15880:
------------------------------------------
I also think we do not have a strong reason to implement this on spark. Since
activities to support new features in this component is too inactive,
maintenance costs are relatively high as compared to gains we get from the
implementation.
> PREGEL Based Semi-Clustering Algorithm Implementation using Spark GraphX API
> ----------------------------------------------------------------------------
>
> Key: SPARK-15880
> URL: https://issues.apache.org/jira/browse/SPARK-15880
> Project: Spark
> Issue Type: New Feature
> Components: GraphX
> Reporter: R J
> Priority: Minor
> Attachments: pregel_paper.pdf
>
> Original Estimate: 672h
> Remaining Estimate: 672h
>
> The main concept of Semi-Clustering algorithm on top of social graphs are:
> - Vertices in a social graph typically represent people, and edges represent
> connections between them.
> - Edges may be based on explicit actions (e.g., adding a friend in a social
> networking site), or may be inferred from people’s behaviour (e.g., email
> conversations or co-publication).
> - Edges may have weights, to represent the interactions frequency or
> strength.
> - A semi-cluster in a social graph is a group of people who interact
> frequently with each other and less frequently with others.
> - What distinguishes it from ordinary clustering is that, a vertex may
> belong to more than one semi-cluster.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
