predictionio-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Pat Ferrel <>
Subject Re: Universal Recommender : seasonality of product
Date Wed, 13 Sep 2017 20:55:45 GMT
You can’t keep data forever or your storage needs will grow without bounds. At some point
you will need to trim it to maintain a moving window in time. Some E-Com applications maintain
a year of data if their items have relatively long lifetimes in the catalog.

You may need to use the db-cleaner template to trim your data to whatever amount you want
to keep. This time-window defines the expiration of the blacklist.

On Sep 13, 2017, at 1:50 PM, Saarthak Chandra <> wrote:

Hi Pat,

Thank you for the reply.
So what do we mean when we say "ages out of the data"?

On Wed, Sep 13, 2017 at 1:42 PM, Pat Ferrel < <>>
This is done with blacklisting. The default config blacklists all items in the training data
that the users has taken the primary event on. So if your primary event is “buy” then
once a user has bought a particular table they will not be recommended that table again until
the “buy” event ages out of the data. If you maintain 1 year of data, then a year after
the user bought the particular table they might get a recommendation for it. In other words
the blacklist expires with the last event in the input data.

On Sep 13, 2017, at 1:36 PM, Saarthak Chandra < <>>


Is there a way to add seasonality of products:
eg: If I buy a table now, I would not want to recommend tables again for the next year. 

So how could I include this seasonality into making recommendations?

Saarthak Chandra,
Masters in Computer Science,
Cornell University.

You received this message because you are subscribed to the Google Groups "actionml-user"
To unsubscribe from this group and stop receiving emails from it, send an email to
To post to this group, send email to <>.
To view this discussion on the web visit
For more options, visit <>.

Saarthak Chandra ,
Masters in Computer Science,
Cornell University.

View raw message