You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2015/06/24 23:52:05 UTC
[jira] [Commented] (FLINK-2271) PageRank gives wrong results with
weighted graph input
[ https://issues.apache.org/jira/browse/FLINK-2271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14600231#comment-14600231 ]
ASF GitHub Bot commented on FLINK-2271:
---------------------------------------
GitHub user vasia opened a pull request:
https://github.com/apache/flink/pull/865
[FLINK-2271] [FLINK-1522] [gelly] add missing example tests
This PR adds missing Gelly tests for PageRank and MusicProfiles examples. It also changes both vertex-centric and GSA PageRank examples to expect an unweighted edge list as input and to properly assign the transition probabilities as edge weights.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/vasia/flink gelly-example-tests
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/865.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #865
----
commit ac6c8ac63d730fd03125b3798778e76267c59dcb
Author: vasia <va...@apache.org>
Date: 2015-06-24T16:09:13Z
[FLINK-2271] [FLINK-1522] [gelly] changed PageRank example to expect an unweighted edge list
added PageRank and MusicProfiles tests
got rid of unused suppress warning annotations in Graph and JaccardSimilarityMeasure
changed GSAPageRank example to expect an unweighted edge list
----
> PageRank gives wrong results with weighted graph input
> ------------------------------------------------------
>
> Key: FLINK-2271
> URL: https://issues.apache.org/jira/browse/FLINK-2271
> Project: Flink
> Issue Type: Bug
> Components: Gelly
> Affects Versions: 0.9
> Reporter: Vasia Kalavri
> Assignee: Vasia Kalavri
> Fix For: 0.10
>
>
> The current implementation of the PageRank algorithm expects a weighted edge list as input. However, if the edge weight is other than 1.0, this will result in wrong results.
> We should change the library method and corresponding examples (also GSAPageRank) to expect an unweighted graph and compute the transition probabilities correctly.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)