Frank McQuillan commented on MADLIB-1084:

For large graphs I fear we may run into memory related problems in the database 
as Nandish mentions.  That is the main reason why we went with an iterative 
approach for PageRank.  A lot of the academic literature around PageRank and 
PPR describe the matrix approach which is harder to implement in a distributed 

I'd suggest we try to find an example of iterative PPR or think through what 
that might look like.  

One relevant source perhaps:

> Graph - Personalized PageRank
> -----------------------------
>                 Key: MADLIB-1084
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1084
>             Project: Apache MADlib
>          Issue Type: New Feature
>          Components: Module: Graph
>            Reporter: Frank McQuillan
>            Assignee: Himanshu Pandey
>            Priority: Major
>             Fix For: v1.14
> Personalized PageRank which is a variant of regular PageRank.
> Please refer to  
> [http://madlib.apache.org/docs/latest/group__grp__pagerank.html] as a 
> starting point.
> Reference:
>  Neighborhood Formation and Anomaly Detection in Bipartite Graphs
>  [http://www.cs.cmu.edu/~deepay/mywww/papers/icdm05.pdf]

This message was sent by Atlassian JIRA

Reply via email to