[
https://issues.apache.org/jira/browse/FLINK-3879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15280134#comment-15280134
]
Greg Hogan commented on FLINK-3879:
-----------------------------------
Implementations which merely showcase the use of Gelly graph models seem most
appropriate as examples. Do we have examples of inputs which perform better as
GSA vs SG vs Pregel? I am not finding any direct guidance in the documentation
for a user looking to choose between duplicate library algorithms.
FLINK-3879 will be faster unless one of the current models is extended or a new
graph model is created to process out- and in- edges separately in the same
iteration or to allow disabling operators on certain supersteps.
An approximate HITS using delta iterations would be as easy to implement
natively as with GSA. Before accepting such an implementation I would like to
see evidence that performing more approximate iterations converges more quickly
when compared with running fewer bulk iterations.
> Native implementation of HITS algorithm
> ---------------------------------------
>
> Key: FLINK-3879
> URL: https://issues.apache.org/jira/browse/FLINK-3879
> Project: Flink
> Issue Type: New Feature
> Components: Gelly
> Affects Versions: 1.1.0
> Reporter: Greg Hogan
> Assignee: Greg Hogan
> Fix For: 1.1.0
>
>
> Hyperlink-Induced Topic Search (HITS, also "hubs and authorities") is
> presented in [0] and described in [1].
> "[HITS] is a very popular and effective algorithm to rank documents based on
> the link information among a set of documents. The algorithm presumes that a
> good hub is a document that points to many others, and a good authority is a
> document that many documents point to."
> [https://pdfs.semanticscholar.org/a8d7/c7a4c53a9102c4239356f9072ec62ca5e62f.pdf]
> This implementation differs from FLINK-2044 by providing for convergence,
> outputting both hub and authority scores, and completing in half the number
> of iterations.
> [0] http://www.cs.cornell.edu/home/kleinber/auth.pdf
> [1] https://en.wikipedia.org/wiki/HITS_algorithm
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
