You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@storm.apache.org by Eugene <ed...@gmail.com> on 2015/06/24 04:06:08 UTC
Upcoming event: Spotify’s Music Recommendations Lambda Architecture with Storm
Please check this event in New York -
Spotify’s Music Recommendations Lambda Architecture
<http://www.meetup.com/New-York-City-Storm-User-Group/events/223407344/>
organized by New York Storm User Group.
Date: Monday, July 13. Place: Spotfy HQ in New York.
Space are limited, please register.
Abstract:
At Spotify, one of the ways we surface personalized music recommendations
to users is via the Discover page. The recommendations are powered by
training large scale Matrix Factorization models on user listening history.
Batch hadoop jobs run daily to build latent vectors for users, tracks,
artists and albums.
However, this setup doesn’t incorporate intra-day listening history.
Furthermore, new users that log into Spotify and stream music will not
receive recommendations until their second day using the application. To
solve the problem, we leverage Storm and process user listens in real-time;
this allows us to surface recommendations to new users as soon as they
start listening to songs on Spotify. We shall discuss our framework, how we
use Storm and challenges we faced building this system.
Thanks
--
Eugene Dvorkin
Organizer
New York City Apache Storm User Group
<http://www.meetup.com/New-York-City-Storm-User-Group/>
Re: Upcoming event: Spotify’s Music Recommendations Lambda Architecture with Storm
Posted by Aniket Alhat <an...@gmail.com>.
Live stream or later video post will be useful.
On Wed, Jun 24, 2015 at 1:41 PM, Scott C. Cote <sc...@gmail.com> wrote:
> Would one of yawl live stream the event (periscope) or if you have a
> prepared url - I can post the url of the stream to the DFW Data Science
> user group.
>
> SCott
> Scott C. Cote
> scottccote@gmail.com
> 972.672.6484
>
>
>
> On Jun 23, 2015, at 9:06 PM, Eugene <ed...@gmail.com> wrote:
>
> Please check this event in New York -
>
> Spotify’s Music Recommendations Lambda Architecture
> <http://www.meetup.com/New-York-City-Storm-User-Group/events/223407344/>
>
> organized by New York Storm User Group.
> Date: Monday, July 13. Place: Spotfy HQ in New York.
> Space are limited, please register.
>
> Abstract:
> At Spotify, one of the ways we surface personalized music recommendations
> to users is via the Discover page. The recommendations are powered by
> training large scale Matrix Factorization models on user listening history.
> Batch hadoop jobs run daily to build latent vectors for users, tracks,
> artists and albums.
>
> However, this setup doesn’t incorporate intra-day listening history.
> Furthermore, new users that log into Spotify and stream music will not
> receive recommendations until their second day using the application. To
> solve the problem, we leverage Storm and process user listens in real-time;
> this allows us to surface recommendations to new users as soon as they
> start listening to songs on Spotify. We shall discuss our framework, how we
> use Storm and challenges we faced building this system.
>
> Thanks
> --
> Eugene Dvorkin
> Organizer
> New York City Apache Storm User Group
> <http://www.meetup.com/New-York-City-Storm-User-Group/>
>
>
>
>
--
*Aniket Alhat*
Re: Upcoming event: Spotify’s Music Recommendations Lambda Architecture with Storm
Posted by "Scott C. Cote" <sc...@gmail.com>.
Would one of yawl live stream the event (periscope) or if you have a prepared url - I can post the url of the stream to the DFW Data Science user group.
SCott
Scott C. Cote
scottccote@gmail.com
972.672.6484
> On Jun 23, 2015, at 9:06 PM, Eugene <ed...@gmail.com> wrote:
>
> Please check this event in New York -
>
> Spotify’s Music Recommendations Lambda Architecture <http://www.meetup.com/New-York-City-Storm-User-Group/events/223407344/>
>
> organized by New York Storm User Group.
> Date: Monday, July 13. Place: Spotfy HQ in New York.
> Space are limited, please register.
>
> Abstract:
> At Spotify, one of the ways we surface personalized music recommendations to users is via the Discover page. The recommendations are powered by training large scale Matrix Factorization models on user listening history. Batch hadoop jobs run daily to build latent vectors for users, tracks, artists and albums.
>
> However, this setup doesn’t incorporate intra-day listening history. Furthermore, new users that log into Spotify and stream music will not receive recommendations until their second day using the application. To solve the problem, we leverage Storm and process user listens in real-time; this allows us to surface recommendations to new users as soon as they start listening to songs on Spotify. We shall discuss our framework, how we use Storm and challenges we faced building this system.
>
> Thanks
> --
> Eugene Dvorkin
> Organizer
> New York City Apache Storm User Group <http://www.meetup.com/New-York-City-Storm-User-Group/>
>
>