recommendations at senscritique. mixing social and machine learning

47
Recommendations at SensCritique Mixing social and machine learning By Xavier RAMPINO, Mars 2016

Upload: recsysfr

Post on 14-Jan-2017

1.166 views

Category:

Data & Analytics


1 download

TRANSCRIPT

Page 1: Recommendations at SensCritique. Mixing social and machine learning

Recommendations at SensCritique Mixing social and machine learning

By Xavier RAMPINO, Mars 2016

Page 2: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Page 3: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Page 4: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Page 5: Recommendations at SensCritique. Mixing social and machine learning

Entertainment database

What is SensCritique ?

Page 6: Recommendations at SensCritique. Mixing social and machine learning

Movies, TV Shows,

Video Games, Books,

Comics, Music

Entertainment database

What is SensCritique ?

Page 7: Recommendations at SensCritique. Mixing social and machine learning

Movies, TV Shows,

Video Games, Books,

Comics, Music

Entertainment database Virtual library

What is SensCritique ?

Page 8: Recommendations at SensCritique. Mixing social and machine learning

Movies, TV Shows,

Video Games, Books,

Comics, Music

Entertainment database Virtual library

Rate, List,

Review, add to a wish list

What is SensCritique ?

Page 9: Recommendations at SensCritique. Mixing social and machine learning

Movies, TV Shows,

Video Games, Books,

Comics, Music

Entertainment database Virtual library Social network

Rate, List,

Review, add to a wish list

What is SensCritique ?

Page 10: Recommendations at SensCritique. Mixing social and machine learning

Movies, TV Shows,

Video Games, Books,

Comics, Music

Entertainment database Virtual library Social network

Rate, List,

Review, add to a wish list

Users can : Follow each other,

like activities, recommend products

What is SensCritique ?

Page 11: Recommendations at SensCritique. Mixing social and machine learning

Database + Library + Network

What is SensCritique ?

Page 12: Recommendations at SensCritique. Mixing social and machine learning

Database + Library + Network

What is SensCritique ?

=

Page 13: Recommendations at SensCritique. Mixing social and machine learning

Database + Library + Network

What is SensCritique ?

=

Page 14: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Database + Library + Network

Page 15: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Database + Library + Network

Page 16: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Database + Library + Network

Discovery Engine

Page 17: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Discovery tool

Database + Library + Network

Discovery Engine

Page 18: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Discovery tool

Discovery tool

Database + Library + Network

Discovery Engine

Page 19: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

Discovery tool

Discovery tool

Discovery tool

Database + Library + Network

Discovery Engine

Page 20: Recommendations at SensCritique. Mixing social and machine learning

What is SensCritique ?

14M

600K

55M

1,5M

products

registered users

ratings

lists

Page 21: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Page 22: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Page 23: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

• Multimedia Database

Entertainment database

Page 24: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

• Multimedia Database

• Extensive Cinema, TV, and VOD showtimes

Entertainment database

Page 25: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

• Multimedia Database

• Extensive Cinema, TV, and VOD showtimes

• Graph Database

Entertainment database

Page 26: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Key factors for discovery :

• Serendipity • Familiar cues (actors, franchises) • Familiar media (TV, cinema)

Entertainment database

Page 27: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

We have very enthusiastic members

Page 28: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our past recommendation engine stack

Page 29: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our past recommendation engine stack

Ratings

Wish

CSV dump LensKit

MySql

Page 30: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our past recommendation engine stack

Very extensive operation, once a day

Ratings

Wish

CSV dump LensKit

MySql

Page 31: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our current recommendation engine stack

Page 32: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our current recommendation engine stack

Nginx Logs

Web hooks

prediction.io event server Application

Page 33: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Virtual Library

Our current recommendation engine stack

Nginx Logs

Web hooks

prediction.io event server Application

• Live Update • Take product views into account • Live boosting during querying (via Elasticsearch)

Page 34: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Social Network

We use social recommendations at two levels

Page 35: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Social Network

We use social recommendations at two levels

User-User recommendation Live Feed

Direct

Page 36: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Social Network

We use social recommendations at two levels

User-User recommendation Live Feed

Direct

News filteringTops

Aggregated

Page 37: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Once one has chosen people to follow, he gets live update on their entertainment activity

Social NetworkDirect

Page 38: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

He may also be given direct recommendation

Social NetworkDirect

Page 39: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

To allow one member to find similar members, we made a user-user recommendation engine using Mahout :

Social NetworkDirect

Online recommendation system, webservice served over Glassfish Event updates are sent on RabbitMQ, then dispatched to a Java Consumer

Page 40: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

One member can notice than a subgroup of member he follows did an activity on a specific product

Social NetworkAggregated

Page 41: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Community choices, given an editorial theme Their own activity (we’ll show them only polls for which

they already rated some of the result products)

Social NetworkAggregated

We have also polls, in which all members can participate. Member then get recommendations fuelled by two forces :

Page 42: Recommendations at SensCritique. Mixing social and machine learning

How we make it work

Social NetworkAggregated

Page 43: Recommendations at SensCritique. Mixing social and machine learning

Conclusion

Page 44: Recommendations at SensCritique. Mixing social and machine learning

From its origins, SensCritique is headed towards product discovery, as a whole though 3 ways :

• Database : We tend to have a comprehensive and ambitious graph-oriented database

• Library : We are a tool for the user to keep a memory of his entertainment history

• Social : We foster social activity between our members, and we are proud of our community

We strongly believe that this is thanks to these three factors that we can offer both a tailored and a global recommendation offer to our audience.

Page 45: Recommendations at SensCritique. Mixing social and machine learning
Page 46: Recommendations at SensCritique. Mixing social and machine learning

Questions ?

Question No question

Page 47: Recommendations at SensCritique. Mixing social and machine learning