Tarek Auel created SPARK-7966:
---------------------------------

             Summary: add Spreading Activation algorithm to GraphX
                 Key: SPARK-7966
                 URL: https://issues.apache.org/jira/browse/SPARK-7966
             Project: Spark
          Issue Type: Improvement
          Components: GraphX
            Reporter: Tarek Auel
            Priority: Minor


I'm wondering if you would like to add the Spreading Activation algorithm to 
GraphX. I have implemented it, using the Pregel-API and would love to share it 
with the community.

Spreading activation is a algorithm that was invented to search in associative 
networks. The basic idea is, that you have one (or multiple) starting nodes. 
The activation spreads out from these nodes to the neighbours and the 
neighbours of the neighbours. The activation decreases after every hop. Nodes 
that were reached by many activations will have a higher total activation level.

Spreading Activation is for many use cases useful. Imagine you have the social 
network of two people. If you apply the spreading activation to this social 
graph with the two people as starting nodes, you will get the nodes that are 
most important for both.

Some resources:

http://www.websci11.org/fileadmin/websci/posters/105_paper.pdf
https://webfiles.uci.edu/eloftus/CollinsLoftus_PsychReview_75.pdf?uniq=20ou4w



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org

Reply via email to