You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@bigtop.apache.org by "jay vyas (JIRA)" <ji...@apache.org> on 2014/11/20 01:43:34 UTC

[jira] [Updated] (BIGTOP-1538) Add customer clustering example to BPS Spark

     [ https://issues.apache.org/jira/browse/BIGTOP-1538?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

jay vyas updated BIGTOP-1538:
-----------------------------
    Assignee: RJ Nowling

> Add customer clustering example to BPS Spark
> --------------------------------------------
>
>                 Key: BIGTOP-1538
>                 URL: https://issues.apache.org/jira/browse/BIGTOP-1538
>             Project: Bigtop
>          Issue Type: Improvement
>          Components: blueprints
>            Reporter: RJ Nowling
>            Assignee: RJ Nowling
>
> We should add an example for how to cluster customers based on purchasing habits.
> Possible approaches for generating vectors:
> 1. Calculate probabilities of purchasing same item, brands, flavors, and quantities repeatedly and cluster vector of probabilities.  (Customer transactions need to be sorted by date / time first)
> 2. For each customer, create vector of normalized purchase counts for each product.  



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