SlideShare a Scribd company logo
Hossein Taghavi
With: Ashok Chandrashekar, Linas Baltrunas, and Justin Basilico
Balancing Discovery and Continuation
in Recommendations
RecSysTV 2016
Outline
§ Background: Netflix recommendations
§ Recommending for different modes of watching
§ Case study: Continue Watching row
§ Conclusions
2
Evolution of Netflix
2006 2016
Netflix Scale
§ > 83M members
§ > 190 countries
§ > 1000 device types
§ > 3.7B hours of content
streamed every month
§ 36% of peak US
downstream traffic
4
§ Recommendations through
predicted star rating
§ Contest:
§ Accuracy measured by root
mean squared error (RMSE)
§ Improve by 10% = $1 million!
§ Data size:
§ 100M ratings (back then
“almost massive”)
5
Turn on Netflix, and the
absolute best contents for you
would automatically start playing
Recommendation System: Ideal State
6
Create a page of recommendations
where the titles you are
most likely to watch and enjoy are
shown on the most visible parts of
the page
Meanwhile…
7
Title Ranking
Everything is a RecommendationRowSelection&Ordering
Recommendations are
driven by machine
learning algorithms
Over 80% of what
members watch comes
from our
recommendations
8
How the Homepage is Built
§ The titles are organized as rows
§ Ordering of titles within rows depends on the row type
§ Selection and ordering of rows:
§ Personalized page generation
algorithm
§ Also some business rules and
constraints
§ Balance thematic coherence,
relevance, and diversity
9
Various Types of Member Interactions/Feedback
§ Plays
§ How long, pause, rewind, skip, etc.
§ Rating and social
§ Rate, like, share
§ Context
§ Time, location, device, language
§ Interactions
§ Scrolling, opening a title page,
search, list add 10
Building the Recommendations is Data Driven
§ Try an idea offline using historical
data to see if it would have made
better recommendations
§ Offline metrics: AUC, nDCG, Recall, …
§ If it did, deploy a live A/B test to see
if it performs well in Production
§ Primary metric: Member retention
Idea /
Problem
Data
Algorithm
Model
Metrics
A/B
Testing
11
For More Reading
§ Netflix tech blog:
§ bit.ly/beyondfivestars
§ bit.ly/learnapage
§ bit.ly/sparktimetravel
12
Building recommendation algorithms that are
balanced for different modes of watching
13
The same you watched last time!
What Is the Most Likely Title You Will Watch?
§ A large portion of watching hours are spent in continue
watching mode
14
Different Modes of Watching
§ Continuation: Resume a
recently-watched TV/Movie
§ List: Play a title previously
added to My List
§ Rewatch: Rewatch a title
enjoyed in the past
§ Discovery: Discover a new
title to watch
15
Recommending for Different Modes:
Approach 1
§ Build one unified model for ranking the titles in each row
and one for ranking rows
§ Optimized for the likelihood of play/enjoyment from the page
§ Benefits:
§ Fewer models to maintain
§ Fewer A/B tests
16
Approach 1: Challenges
§ Members behave differently in different modes
§ Different row types are designed for different behaviors
§ Hard to capture and balance all that in one objective
§ E.g. simply ranking titles by likelihood of play will fill the page with
already-watched titles è Poor member experience
§ Recommendations for different modes have different
sensitivities to member actions
§ Continuation recs may react immediately to watching activities,
My List recs may react to My List add/remove activities, etc.
17
Approach 2: Dedicated Models + Blend
§ Build separate models for the each mode
§ Blend the results on the page
§ Blending can be done through a model trained offline, or a
parameter tuned online
§ E.g., one or more dedicated rows for each mode
§ Pro:
§ More modular, provides more intuitive knobs for balancing
§ Con:
§ Less elegant, more maintenance 18
Case Study: Continue Watching Row
19
Continue Watching Row: The Past
§ CW row was shown on some devices
§ Videos sorted by recency of last watch
§ Row appearance on page by business rules
§ On the website, only a single CW title
§ A very significant fraction of plays are continuations
§ CW deserved a better treatment
20
Objective
§ Unify the CW row across devices
§ Optimize the row in two dimensions:
§ Row position on page
§ Place it higher when the member is more
likely to resume a video
§ Re-order the titles within the CW row
§ By their likelihood to be resumed in the
current session
21
Some Intuitive Patterns
§ Member may be more likely to want to
§ Resume a video if:
§ In the middle of binging a TV show
§ Partially watched a movie recently
§ Often watched it around this time of the day, location, or on the current
device
§ Discover a new title if:
§ Just finished a movie or completed all episodes of a show
§ Hasn’t watched anything recently
§ Is a relatively new member
22
Building a Recommendation Model for CW
§ Feature Brainstorm
§ Training Data
§ Models and Metrics
§ Implementation
23
Feature Ideas
§ Member-level:
§ Member’s subscription: tenure, country, language
§ How active has the member been recently
§ Member past ratings, genre preferences, etc.
24
Feature Ideas
§ Video and member’s previous interactions with it:
§ How recently was the video added to the catalog, watched, ...
§ How much of the movie/show watched
§ Video metadata:
§ Type and genre of video, # episodes
§ E.g., kids titles may be re-watched more
§ What else is on the catalog
§ Popularity and relevance of the video
§ How often do members resume this video
25
Feature Ideas
§ Contextual:
§ Time of the day and day of the week
§ Location at various resolutions
§ Device
26
Title Ranking Model
§ Training data
§ Continuation sessions
§ Look at which of the recently-watched titles were played?
§ Model
§ Learn-to-rank: Linear/ensembles/…
§ Optimize for how well we rank the played title among other titles
27
Title Ranking Model: Performance
§ Baseline: Ranking by recency of
last play
§ Recency rank was also an
important feature in the model
§ Metrics significantly higher than
the baseline
§ E.g. Significant lift in precision
§ A/B testing also showed
improvements
28
Row Placement Model
§ Objective
§ Estimate the likelihood of continuation vs. discovery
§ Map that likelihood to a position on the page
§ Simplification:
§ Fix two candidate positions on the page and apply a threshold
§ Tune the threshold to optimize some accuracy metric
29
Row Placement Model: Training
§ Training data
§ Randomly select sessions with plays globally
§ Model
§ Binary classification of continuation vs. discovery sessions
§ Evaluated using classification and ranking metrics
30
Row Placement Model: Performance
§ Metrics
§ Achieved high classification metrics for predicting continuation vs
discovery
§ Error types:
§ False positives è CW occupies top of the page unnecessarily
§ False negative è Difficult for member to find the CW title
§ Placing the row
§ Threshold trades off FP and FN è Hard to tune offline
§ Tuned the threshold by A/B testing
31
Reusing the Title Ranking Model
§ Use the title-level scores
§ Calibrate scores to get probability Pt of continuation for each CW
title t
§ Aggregate into an overall probability of continuation
§ E.g., assuming independence:
PCW = 1 - ∏tϵCW (1- Pt)
§ Pro: Avoids maintaining two separate models
§ Con: Not as accurate as a dedicated model
32
Context Awareness
§ Title ranks highest on the same time of day and device
as last play
§ Experiment:
§ Played “Sid the Science Kid” on iPhone
§ Played “Narcos” on the website
è Different ranking on iPhone and Web
33
Serving the CW Row in Production
§ Score cannot be precomputed è Real- or near real-time
§ Some features are context dependent
§ Row should refresh each time a member watches a title
§ Need to push updates to clients to keep the row fresh
§ Latency bottleneck: Data transfers from the cache to
computation backend
§ Requires careful backend engineering
§ Fallback strategy: If computation fails, can use recency ranking
34
Conclusions and Future Directions
35
Conclusions
§ Important to understand different modes of behavior
§ Continuation is a key driver of streaming hours
§ Improving CW recommendations improves member experience
§ A/B testing showed significant boost in user engagement
§ Future:
§ Incorporate the placement of CW row (and others) into the main
page construction model
§ When can we automatically start resuming a title? 36
Questions?
Upcoming blog post on this topic at: techblog.netflix.com
Job openings: jobs.netflix.com
37

More Related Content

Viewers also liked

(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning
Yves Raimond
 
Time series effects for TV recommendations
Time series effects for TV recommendationsTime series effects for TV recommendations
Time series effects for TV recommendations
IntoTheMinds
 
Datafying public sphere : fragmented audience , ùmedia and democracy
Datafying public sphere : fragmented audience , ùmedia and democracyDatafying public sphere : fragmented audience , ùmedia and democracy
Datafying public sphere : fragmented audience , ùmedia and democracy
IntoTheMinds
 
Personalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing RecommendationsPersonalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing Recommendations
Justin Basilico
 
Big & Personal: the data and the models behind Netflix recommendations by Xa...
 Big & Personal: the data and the models behind Netflix recommendations by Xa... Big & Personal: the data and the models behind Netflix recommendations by Xa...
Big & Personal: the data and the models behind Netflix recommendations by Xa...
BigMine
 
Recommendation at Netflix Scale
Recommendation at Netflix ScaleRecommendation at Netflix Scale
Recommendation at Netflix Scale
Justin Basilico
 
9 17-16 - when recommendation systems go bad - rec sys
9 17-16 - when recommendation systems go bad - rec sys9 17-16 - when recommendation systems go bad - rec sys
9 17-16 - when recommendation systems go bad - rec sys
Evan Estola
 
Personalization - 10 Lessons Learned from Netflix
Personalization - 10 Lessons Learned from NetflixPersonalization - 10 Lessons Learned from Netflix
Personalization - 10 Lessons Learned from Netflix
Pancrazio Auteri
 
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.jsNetflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
Chris Saint-Amant
 
Culture
CultureCulture
Culture
Reed Hastings
 

Viewers also liked (10)

(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning(Some) pitfalls of distributed learning
(Some) pitfalls of distributed learning
 
Time series effects for TV recommendations
Time series effects for TV recommendationsTime series effects for TV recommendations
Time series effects for TV recommendations
 
Datafying public sphere : fragmented audience , ùmedia and democracy
Datafying public sphere : fragmented audience , ùmedia and democracyDatafying public sphere : fragmented audience , ùmedia and democracy
Datafying public sphere : fragmented audience , ùmedia and democracy
 
Personalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing RecommendationsPersonalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing Recommendations
 
Big & Personal: the data and the models behind Netflix recommendations by Xa...
 Big & Personal: the data and the models behind Netflix recommendations by Xa... Big & Personal: the data and the models behind Netflix recommendations by Xa...
Big & Personal: the data and the models behind Netflix recommendations by Xa...
 
Recommendation at Netflix Scale
Recommendation at Netflix ScaleRecommendation at Netflix Scale
Recommendation at Netflix Scale
 
9 17-16 - when recommendation systems go bad - rec sys
9 17-16 - when recommendation systems go bad - rec sys9 17-16 - when recommendation systems go bad - rec sys
9 17-16 - when recommendation systems go bad - rec sys
 
Personalization - 10 Lessons Learned from Netflix
Personalization - 10 Lessons Learned from NetflixPersonalization - 10 Lessons Learned from Netflix
Personalization - 10 Lessons Learned from Netflix
 
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.jsNetflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
Netflix JavaScript Talks - Scaling A/B Testing on Netflix.com with Node.js
 
Culture
CultureCulture
Culture
 

Similar to Balancing Discovery and continuation in recommendation (hossein taghavi netflix)

Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix Scale
Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix ScaleQcon SF 2013 - Machine Learning & Recommender Systems @ Netflix Scale
Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix ScaleXavier Amatriain
 
SocialRating
SocialRatingSocialRating
SocialRating
Rohit G
 
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, EuropePatterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
Flip Kromer
 
Morph your mindset for Continuous Delivery, Agile Roots 2014
Morph your mindset for Continuous Delivery,  Agile Roots 2014Morph your mindset for Continuous Delivery,  Agile Roots 2014
Morph your mindset for Continuous Delivery, Agile Roots 2014
lisacrispin
 
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
SOASTA
 
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
DataScienceConferenc1
 
Video Recommendation Engines as a Service
Video Recommendation Engines as a ServiceVideo Recommendation Engines as a Service
Video Recommendation Engines as a Service
Kamil Sindi
 
Project 5 E-Commerce
Project 5   E-CommerceProject 5   E-Commerce
Project 5 E-Commerce
Denese Doby
 
Majestic Workshop on Backlinks and Link Building
Majestic Workshop on Backlinks and Link BuildingMajestic Workshop on Backlinks and Link Building
Majestic Workshop on Backlinks and Link Building
Sante J. Achille
 
Advanced Testing and Debugging using the Developer Console webinar
Advanced Testing and Debugging using the Developer Console webinarAdvanced Testing and Debugging using the Developer Console webinar
Advanced Testing and Debugging using the Developer Console webinar
Salesforce Developers
 
Socialtext Distributed Agile
Socialtext Distributed AgileSocialtext Distributed Agile
Socialtext Distributed Agile
Ross Mayfield
 
Lean Kanban India 2016 | Stealth Kanban | Kirill Klimov
Lean Kanban India 2016 | Stealth Kanban | Kirill KlimovLean Kanban India 2016 | Stealth Kanban | Kirill Klimov
Lean Kanban India 2016 | Stealth Kanban | Kirill Klimov
LeanKanbanIndia
 
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
Christopher Hamm
 
Pega Lead System Architecture (CPLSA) Exam | Start Your Preparation
Pega Lead System Architecture (CPLSA) Exam | Start Your PreparationPega Lead System Architecture (CPLSA) Exam | Start Your Preparation
Pega Lead System Architecture (CPLSA) Exam | Start Your Preparation
Meghna Arora
 
MongoDB.local Seattle 2019: Advanced Schema Design Patterns
MongoDB.local Seattle 2019: Advanced Schema Design PatternsMongoDB.local Seattle 2019: Advanced Schema Design Patterns
MongoDB.local Seattle 2019: Advanced Schema Design Patterns
MongoDB
 
Spec by-example
Spec by-exampleSpec by-example
Spec by-example
David Navarro Alvarez
 
Stealth Kanban
Stealth KanbanStealth Kanban
Stealth Kanban
Kirill Klimov
 
MongoDB.local Dallas 2019: Advanced Schema Design Patterns
MongoDB.local Dallas 2019: Advanced Schema Design PatternsMongoDB.local Dallas 2019: Advanced Schema Design Patterns
MongoDB.local Dallas 2019: Advanced Schema Design Patterns
MongoDB
 
National Geographic Competitive Assessment
National Geographic Competitive AssessmentNational Geographic Competitive Assessment
National Geographic Competitive Assessment
Nelson Gaytón
 
Supporting studio-based design courses with django-courseapp
Supporting studio-based design courses with django-courseappSupporting studio-based design courses with django-courseapp
Supporting studio-based design courses with django-courseapp
Mike Krieger
 

Similar to Balancing Discovery and continuation in recommendation (hossein taghavi netflix) (20)

Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix Scale
Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix ScaleQcon SF 2013 - Machine Learning & Recommender Systems @ Netflix Scale
Qcon SF 2013 - Machine Learning & Recommender Systems @ Netflix Scale
 
SocialRating
SocialRatingSocialRating
SocialRating
 
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, EuropePatterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
Patterns of the Lambda Architecture -- 2015 April - Hadoop Summit, Europe
 
Morph your mindset for Continuous Delivery, Agile Roots 2014
Morph your mindset for Continuous Delivery,  Agile Roots 2014Morph your mindset for Continuous Delivery,  Agile Roots 2014
Morph your mindset for Continuous Delivery, Agile Roots 2014
 
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
Keys To World-Class Retail Web Performance - Expert tips for holiday web read...
 
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
[DSC Europe 23] Vladislav Belov - ChatBot Learning Assistant with Large Langu...
 
Video Recommendation Engines as a Service
Video Recommendation Engines as a ServiceVideo Recommendation Engines as a Service
Video Recommendation Engines as a Service
 
Project 5 E-Commerce
Project 5   E-CommerceProject 5   E-Commerce
Project 5 E-Commerce
 
Majestic Workshop on Backlinks and Link Building
Majestic Workshop on Backlinks and Link BuildingMajestic Workshop on Backlinks and Link Building
Majestic Workshop on Backlinks and Link Building
 
Advanced Testing and Debugging using the Developer Console webinar
Advanced Testing and Debugging using the Developer Console webinarAdvanced Testing and Debugging using the Developer Console webinar
Advanced Testing and Debugging using the Developer Console webinar
 
Socialtext Distributed Agile
Socialtext Distributed AgileSocialtext Distributed Agile
Socialtext Distributed Agile
 
Lean Kanban India 2016 | Stealth Kanban | Kirill Klimov
Lean Kanban India 2016 | Stealth Kanban | Kirill KlimovLean Kanban India 2016 | Stealth Kanban | Kirill Klimov
Lean Kanban India 2016 | Stealth Kanban | Kirill Klimov
 
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
Code Palousa presentation- "Giving Digital Eyes to your Synthetic Tests"
 
Pega Lead System Architecture (CPLSA) Exam | Start Your Preparation
Pega Lead System Architecture (CPLSA) Exam | Start Your PreparationPega Lead System Architecture (CPLSA) Exam | Start Your Preparation
Pega Lead System Architecture (CPLSA) Exam | Start Your Preparation
 
MongoDB.local Seattle 2019: Advanced Schema Design Patterns
MongoDB.local Seattle 2019: Advanced Schema Design PatternsMongoDB.local Seattle 2019: Advanced Schema Design Patterns
MongoDB.local Seattle 2019: Advanced Schema Design Patterns
 
Spec by-example
Spec by-exampleSpec by-example
Spec by-example
 
Stealth Kanban
Stealth KanbanStealth Kanban
Stealth Kanban
 
MongoDB.local Dallas 2019: Advanced Schema Design Patterns
MongoDB.local Dallas 2019: Advanced Schema Design PatternsMongoDB.local Dallas 2019: Advanced Schema Design Patterns
MongoDB.local Dallas 2019: Advanced Schema Design Patterns
 
National Geographic Competitive Assessment
National Geographic Competitive AssessmentNational Geographic Competitive Assessment
National Geographic Competitive Assessment
 
Supporting studio-based design courses with django-courseapp
Supporting studio-based design courses with django-courseappSupporting studio-based design courses with django-courseapp
Supporting studio-based design courses with django-courseapp
 

More from IntoTheMinds

Voilà à quoi ressemblera la reprise
Voilà à quoi ressemblera la repriseVoilà à quoi ressemblera la reprise
Voilà à quoi ressemblera la reprise
IntoTheMinds
 
The advertising campaigns run in Belgium during the Covid-19 crisis
The advertising campaigns run in Belgium during the Covid-19 crisisThe advertising campaigns run in Belgium during the Covid-19 crisis
The advertising campaigns run in Belgium during the Covid-19 crisis
IntoTheMinds
 
Presentation Christian Radler at EBU Conference "data in the newsroom"
Presentation Christian Radler at EBU Conference "data in the newsroom"Presentation Christian Radler at EBU Conference "data in the newsroom"
Presentation Christian Radler at EBU Conference "data in the newsroom"
IntoTheMinds
 
Presentation Sabino Metta at EBU Conference "data in the newsroom"
Presentation Sabino Metta at EBU Conference "data in the newsroom"Presentation Sabino Metta at EBU Conference "data in the newsroom"
Presentation Sabino Metta at EBU Conference "data in the newsroom"
IntoTheMinds
 
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
IntoTheMinds
 
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
IntoTheMinds
 
Purchase drivers for iconic products in the luxury sector
Purchase drivers for iconic products in the luxury sectorPurchase drivers for iconic products in the luxury sector
Purchase drivers for iconic products in the luxury sector
IntoTheMinds
 
A robot called Voitto
A robot called VoittoA robot called Voitto
A robot called Voitto
IntoTheMinds
 
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
IntoTheMinds
 
Artworks personalization on Netflix
Artworks personalization on Netflix Artworks personalization on Netflix
Artworks personalization on Netflix
IntoTheMinds
 
Privacy Calculus in the Sharing Economy
Privacy Calculus in the Sharing EconomyPrivacy Calculus in the Sharing Economy
Privacy Calculus in the Sharing Economy
IntoTheMinds
 
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
IntoTheMinds
 
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
IntoTheMinds
 
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
IntoTheMinds
 
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
IntoTheMinds
 
Big Data and ethics meetup : slides presentation michael ekstrand
Big Data and ethics meetup : slides presentation michael ekstrandBig Data and ethics meetup : slides presentation michael ekstrand
Big Data and ethics meetup : slides presentation michael ekstrand
IntoTheMinds
 
Presentatie big data (Dag van de verkoper, Cevora)
Presentatie big data (Dag van de verkoper, Cevora) Presentatie big data (Dag van de verkoper, Cevora)
Presentatie big data (Dag van de verkoper, Cevora)
IntoTheMinds
 
Presentatie big data in verkoop (cevora) gent 16 Mei 2017
Presentatie big data in verkoop (cevora) gent 16 Mei 2017Presentatie big data in verkoop (cevora) gent 16 Mei 2017
Presentatie big data in verkoop (cevora) gent 16 Mei 2017
IntoTheMinds
 
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
IntoTheMinds
 
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic..."Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
IntoTheMinds
 

More from IntoTheMinds (20)

Voilà à quoi ressemblera la reprise
Voilà à quoi ressemblera la repriseVoilà à quoi ressemblera la reprise
Voilà à quoi ressemblera la reprise
 
The advertising campaigns run in Belgium during the Covid-19 crisis
The advertising campaigns run in Belgium during the Covid-19 crisisThe advertising campaigns run in Belgium during the Covid-19 crisis
The advertising campaigns run in Belgium during the Covid-19 crisis
 
Presentation Christian Radler at EBU Conference "data in the newsroom"
Presentation Christian Radler at EBU Conference "data in the newsroom"Presentation Christian Radler at EBU Conference "data in the newsroom"
Presentation Christian Radler at EBU Conference "data in the newsroom"
 
Presentation Sabino Metta at EBU Conference "data in the newsroom"
Presentation Sabino Metta at EBU Conference "data in the newsroom"Presentation Sabino Metta at EBU Conference "data in the newsroom"
Presentation Sabino Metta at EBU Conference "data in the newsroom"
 
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
Presentation Stéphane Saulnier at EBU Conference "data in the newsroom"
 
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
Presentation Kristofer Sjoholm at EBU Conference "data in the newsroom"
 
Purchase drivers for iconic products in the luxury sector
Purchase drivers for iconic products in the luxury sectorPurchase drivers for iconic products in the luxury sector
Purchase drivers for iconic products in the luxury sector
 
A robot called Voitto
A robot called VoittoA robot called Voitto
A robot called Voitto
 
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
presentation Newsbridge by Philippe Petitpont at Media Fast Forward 2018
 
Artworks personalization on Netflix
Artworks personalization on Netflix Artworks personalization on Netflix
Artworks personalization on Netflix
 
Privacy Calculus in the Sharing Economy
Privacy Calculus in the Sharing EconomyPrivacy Calculus in the Sharing Economy
Privacy Calculus in the Sharing Economy
 
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
Toon borré presentation at Meetup Big Data and Ethics at DigitYser Brussels 1...
 
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
Leenke De Donder presentation at Meetup Big Data and Ethics at DigitYser Brus...
 
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
Jochanen eynikel presentation at Meetup Big Data and Ethics at DigitYser Brus...
 
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
Thomas carette presentation at Meetup Big Data and Ethics at DigitYser Brusse...
 
Big Data and ethics meetup : slides presentation michael ekstrand
Big Data and ethics meetup : slides presentation michael ekstrandBig Data and ethics meetup : slides presentation michael ekstrand
Big Data and ethics meetup : slides presentation michael ekstrand
 
Presentatie big data (Dag van de verkoper, Cevora)
Presentatie big data (Dag van de verkoper, Cevora) Presentatie big data (Dag van de verkoper, Cevora)
Presentatie big data (Dag van de verkoper, Cevora)
 
Presentatie big data in verkoop (cevora) gent 16 Mei 2017
Presentatie big data in verkoop (cevora) gent 16 Mei 2017Presentatie big data in verkoop (cevora) gent 16 Mei 2017
Presentatie big data in verkoop (cevora) gent 16 Mei 2017
 
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
Slides pierre nicolas schwab DISummit 2017 (Big Data, Brussels)
 
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic..."Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
"Building Trust" discussion panel at EBU Big Data conference 2017 (Pierre-Nic...
 

Recently uploaded

Pitch Deck Teardown: RAW Dating App's $3M Angel deck
Pitch Deck Teardown: RAW Dating App's $3M Angel deckPitch Deck Teardown: RAW Dating App's $3M Angel deck
Pitch Deck Teardown: RAW Dating App's $3M Angel deck
HajeJanKamps
 
Brand Analysis for an artist named Struan
Brand Analysis for an artist named StruanBrand Analysis for an artist named Struan
Brand Analysis for an artist named Struan
sarahvanessa51503
 
Unveiling the Secrets How Does Generative AI Work.pdf
Unveiling the Secrets How Does Generative AI Work.pdfUnveiling the Secrets How Does Generative AI Work.pdf
Unveiling the Secrets How Does Generative AI Work.pdf
Sam H
 
Improving profitability for small business
Improving profitability for small businessImproving profitability for small business
Improving profitability for small business
Ben Wann
 
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
BBPMedia1
 
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
Kumar Satyam
 
Discover the innovative and creative projects that highlight my journey throu...
Discover the innovative and creative projects that highlight my journey throu...Discover the innovative and creative projects that highlight my journey throu...
Discover the innovative and creative projects that highlight my journey throu...
dylandmeas
 
Buy Verified PayPal Account | Buy Google 5 Star Reviews
Buy Verified PayPal Account | Buy Google 5 Star ReviewsBuy Verified PayPal Account | Buy Google 5 Star Reviews
Buy Verified PayPal Account | Buy Google 5 Star Reviews
usawebmarket
 
Business Valuation Principles for Entrepreneurs
Business Valuation Principles for EntrepreneursBusiness Valuation Principles for Entrepreneurs
Business Valuation Principles for Entrepreneurs
Ben Wann
 
Global Interconnection Group Joint Venture[960] (1).pdf
Global Interconnection Group Joint Venture[960] (1).pdfGlobal Interconnection Group Joint Venture[960] (1).pdf
Global Interconnection Group Joint Venture[960] (1).pdf
Henry Tapper
 
Set off and carry forward of losses and assessment of individuals.pptx
Set off and carry forward of losses and assessment of individuals.pptxSet off and carry forward of losses and assessment of individuals.pptx
Set off and carry forward of losses and assessment of individuals.pptx
HARSHITHV26
 
5 Things You Need To Know Before Hiring a Videographer
5 Things You Need To Know Before Hiring a Videographer5 Things You Need To Know Before Hiring a Videographer
5 Things You Need To Know Before Hiring a Videographer
ofm712785
 
Skye Residences | Extended Stay Residences Near Toronto Airport
Skye Residences | Extended Stay Residences Near Toronto AirportSkye Residences | Extended Stay Residences Near Toronto Airport
Skye Residences | Extended Stay Residences Near Toronto Airport
marketingjdass
 
BeMetals Presentation_May_22_2024 .pdf
BeMetals Presentation_May_22_2024   .pdfBeMetals Presentation_May_22_2024   .pdf
BeMetals Presentation_May_22_2024 .pdf
DerekIwanaka1
 
The Parable of the Pipeline a book every new businessman or business student ...
The Parable of the Pipeline a book every new businessman or business student ...The Parable of the Pipeline a book every new businessman or business student ...
The Parable of the Pipeline a book every new businessman or business student ...
awaisafdar
 
falcon-invoice-discounting-a-premier-platform-for-investors-in-india
falcon-invoice-discounting-a-premier-platform-for-investors-in-indiafalcon-invoice-discounting-a-premier-platform-for-investors-in-india
falcon-invoice-discounting-a-premier-platform-for-investors-in-india
Falcon Invoice Discounting
 
Cree_Rey_BrandIdentityKit.PDF_PersonalBd
Cree_Rey_BrandIdentityKit.PDF_PersonalBdCree_Rey_BrandIdentityKit.PDF_PersonalBd
Cree_Rey_BrandIdentityKit.PDF_PersonalBd
creerey
 
Memorandum Of Association Constitution of Company.ppt
Memorandum Of Association Constitution of Company.pptMemorandum Of Association Constitution of Company.ppt
Memorandum Of Association Constitution of Company.ppt
seri bangash
 
The-McKinsey-7S-Framework. strategic management
The-McKinsey-7S-Framework. strategic managementThe-McKinsey-7S-Framework. strategic management
The-McKinsey-7S-Framework. strategic management
Bojamma2
 
Affordable Stationery Printing Services in Jaipur | Navpack n Print
Affordable Stationery Printing Services in Jaipur | Navpack n PrintAffordable Stationery Printing Services in Jaipur | Navpack n Print
Affordable Stationery Printing Services in Jaipur | Navpack n Print
Navpack & Print
 

Recently uploaded (20)

Pitch Deck Teardown: RAW Dating App's $3M Angel deck
Pitch Deck Teardown: RAW Dating App's $3M Angel deckPitch Deck Teardown: RAW Dating App's $3M Angel deck
Pitch Deck Teardown: RAW Dating App's $3M Angel deck
 
Brand Analysis for an artist named Struan
Brand Analysis for an artist named StruanBrand Analysis for an artist named Struan
Brand Analysis for an artist named Struan
 
Unveiling the Secrets How Does Generative AI Work.pdf
Unveiling the Secrets How Does Generative AI Work.pdfUnveiling the Secrets How Does Generative AI Work.pdf
Unveiling the Secrets How Does Generative AI Work.pdf
 
Improving profitability for small business
Improving profitability for small businessImproving profitability for small business
Improving profitability for small business
 
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
RMD24 | Debunking the non-endemic revenue myth Marvin Vacquier Droop | First ...
 
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
India Orthopedic Devices Market: Unlocking Growth Secrets, Trends and Develop...
 
Discover the innovative and creative projects that highlight my journey throu...
Discover the innovative and creative projects that highlight my journey throu...Discover the innovative and creative projects that highlight my journey throu...
Discover the innovative and creative projects that highlight my journey throu...
 
Buy Verified PayPal Account | Buy Google 5 Star Reviews
Buy Verified PayPal Account | Buy Google 5 Star ReviewsBuy Verified PayPal Account | Buy Google 5 Star Reviews
Buy Verified PayPal Account | Buy Google 5 Star Reviews
 
Business Valuation Principles for Entrepreneurs
Business Valuation Principles for EntrepreneursBusiness Valuation Principles for Entrepreneurs
Business Valuation Principles for Entrepreneurs
 
Global Interconnection Group Joint Venture[960] (1).pdf
Global Interconnection Group Joint Venture[960] (1).pdfGlobal Interconnection Group Joint Venture[960] (1).pdf
Global Interconnection Group Joint Venture[960] (1).pdf
 
Set off and carry forward of losses and assessment of individuals.pptx
Set off and carry forward of losses and assessment of individuals.pptxSet off and carry forward of losses and assessment of individuals.pptx
Set off and carry forward of losses and assessment of individuals.pptx
 
5 Things You Need To Know Before Hiring a Videographer
5 Things You Need To Know Before Hiring a Videographer5 Things You Need To Know Before Hiring a Videographer
5 Things You Need To Know Before Hiring a Videographer
 
Skye Residences | Extended Stay Residences Near Toronto Airport
Skye Residences | Extended Stay Residences Near Toronto AirportSkye Residences | Extended Stay Residences Near Toronto Airport
Skye Residences | Extended Stay Residences Near Toronto Airport
 
BeMetals Presentation_May_22_2024 .pdf
BeMetals Presentation_May_22_2024   .pdfBeMetals Presentation_May_22_2024   .pdf
BeMetals Presentation_May_22_2024 .pdf
 
The Parable of the Pipeline a book every new businessman or business student ...
The Parable of the Pipeline a book every new businessman or business student ...The Parable of the Pipeline a book every new businessman or business student ...
The Parable of the Pipeline a book every new businessman or business student ...
 
falcon-invoice-discounting-a-premier-platform-for-investors-in-india
falcon-invoice-discounting-a-premier-platform-for-investors-in-indiafalcon-invoice-discounting-a-premier-platform-for-investors-in-india
falcon-invoice-discounting-a-premier-platform-for-investors-in-india
 
Cree_Rey_BrandIdentityKit.PDF_PersonalBd
Cree_Rey_BrandIdentityKit.PDF_PersonalBdCree_Rey_BrandIdentityKit.PDF_PersonalBd
Cree_Rey_BrandIdentityKit.PDF_PersonalBd
 
Memorandum Of Association Constitution of Company.ppt
Memorandum Of Association Constitution of Company.pptMemorandum Of Association Constitution of Company.ppt
Memorandum Of Association Constitution of Company.ppt
 
The-McKinsey-7S-Framework. strategic management
The-McKinsey-7S-Framework. strategic managementThe-McKinsey-7S-Framework. strategic management
The-McKinsey-7S-Framework. strategic management
 
Affordable Stationery Printing Services in Jaipur | Navpack n Print
Affordable Stationery Printing Services in Jaipur | Navpack n PrintAffordable Stationery Printing Services in Jaipur | Navpack n Print
Affordable Stationery Printing Services in Jaipur | Navpack n Print
 

Balancing Discovery and continuation in recommendation (hossein taghavi netflix)

  • 1. Hossein Taghavi With: Ashok Chandrashekar, Linas Baltrunas, and Justin Basilico Balancing Discovery and Continuation in Recommendations RecSysTV 2016
  • 2. Outline § Background: Netflix recommendations § Recommending for different modes of watching § Case study: Continue Watching row § Conclusions 2
  • 4. Netflix Scale § > 83M members § > 190 countries § > 1000 device types § > 3.7B hours of content streamed every month § 36% of peak US downstream traffic 4
  • 5. § Recommendations through predicted star rating § Contest: § Accuracy measured by root mean squared error (RMSE) § Improve by 10% = $1 million! § Data size: § 100M ratings (back then “almost massive”) 5
  • 6. Turn on Netflix, and the absolute best contents for you would automatically start playing Recommendation System: Ideal State 6
  • 7. Create a page of recommendations where the titles you are most likely to watch and enjoy are shown on the most visible parts of the page Meanwhile… 7
  • 8. Title Ranking Everything is a RecommendationRowSelection&Ordering Recommendations are driven by machine learning algorithms Over 80% of what members watch comes from our recommendations 8
  • 9. How the Homepage is Built § The titles are organized as rows § Ordering of titles within rows depends on the row type § Selection and ordering of rows: § Personalized page generation algorithm § Also some business rules and constraints § Balance thematic coherence, relevance, and diversity 9
  • 10. Various Types of Member Interactions/Feedback § Plays § How long, pause, rewind, skip, etc. § Rating and social § Rate, like, share § Context § Time, location, device, language § Interactions § Scrolling, opening a title page, search, list add 10
  • 11. Building the Recommendations is Data Driven § Try an idea offline using historical data to see if it would have made better recommendations § Offline metrics: AUC, nDCG, Recall, … § If it did, deploy a live A/B test to see if it performs well in Production § Primary metric: Member retention Idea / Problem Data Algorithm Model Metrics A/B Testing 11
  • 12. For More Reading § Netflix tech blog: § bit.ly/beyondfivestars § bit.ly/learnapage § bit.ly/sparktimetravel 12
  • 13. Building recommendation algorithms that are balanced for different modes of watching 13
  • 14. The same you watched last time! What Is the Most Likely Title You Will Watch? § A large portion of watching hours are spent in continue watching mode 14
  • 15. Different Modes of Watching § Continuation: Resume a recently-watched TV/Movie § List: Play a title previously added to My List § Rewatch: Rewatch a title enjoyed in the past § Discovery: Discover a new title to watch 15
  • 16. Recommending for Different Modes: Approach 1 § Build one unified model for ranking the titles in each row and one for ranking rows § Optimized for the likelihood of play/enjoyment from the page § Benefits: § Fewer models to maintain § Fewer A/B tests 16
  • 17. Approach 1: Challenges § Members behave differently in different modes § Different row types are designed for different behaviors § Hard to capture and balance all that in one objective § E.g. simply ranking titles by likelihood of play will fill the page with already-watched titles è Poor member experience § Recommendations for different modes have different sensitivities to member actions § Continuation recs may react immediately to watching activities, My List recs may react to My List add/remove activities, etc. 17
  • 18. Approach 2: Dedicated Models + Blend § Build separate models for the each mode § Blend the results on the page § Blending can be done through a model trained offline, or a parameter tuned online § E.g., one or more dedicated rows for each mode § Pro: § More modular, provides more intuitive knobs for balancing § Con: § Less elegant, more maintenance 18
  • 19. Case Study: Continue Watching Row 19
  • 20. Continue Watching Row: The Past § CW row was shown on some devices § Videos sorted by recency of last watch § Row appearance on page by business rules § On the website, only a single CW title § A very significant fraction of plays are continuations § CW deserved a better treatment 20
  • 21. Objective § Unify the CW row across devices § Optimize the row in two dimensions: § Row position on page § Place it higher when the member is more likely to resume a video § Re-order the titles within the CW row § By their likelihood to be resumed in the current session 21
  • 22. Some Intuitive Patterns § Member may be more likely to want to § Resume a video if: § In the middle of binging a TV show § Partially watched a movie recently § Often watched it around this time of the day, location, or on the current device § Discover a new title if: § Just finished a movie or completed all episodes of a show § Hasn’t watched anything recently § Is a relatively new member 22
  • 23. Building a Recommendation Model for CW § Feature Brainstorm § Training Data § Models and Metrics § Implementation 23
  • 24. Feature Ideas § Member-level: § Member’s subscription: tenure, country, language § How active has the member been recently § Member past ratings, genre preferences, etc. 24
  • 25. Feature Ideas § Video and member’s previous interactions with it: § How recently was the video added to the catalog, watched, ... § How much of the movie/show watched § Video metadata: § Type and genre of video, # episodes § E.g., kids titles may be re-watched more § What else is on the catalog § Popularity and relevance of the video § How often do members resume this video 25
  • 26. Feature Ideas § Contextual: § Time of the day and day of the week § Location at various resolutions § Device 26
  • 27. Title Ranking Model § Training data § Continuation sessions § Look at which of the recently-watched titles were played? § Model § Learn-to-rank: Linear/ensembles/… § Optimize for how well we rank the played title among other titles 27
  • 28. Title Ranking Model: Performance § Baseline: Ranking by recency of last play § Recency rank was also an important feature in the model § Metrics significantly higher than the baseline § E.g. Significant lift in precision § A/B testing also showed improvements 28
  • 29. Row Placement Model § Objective § Estimate the likelihood of continuation vs. discovery § Map that likelihood to a position on the page § Simplification: § Fix two candidate positions on the page and apply a threshold § Tune the threshold to optimize some accuracy metric 29
  • 30. Row Placement Model: Training § Training data § Randomly select sessions with plays globally § Model § Binary classification of continuation vs. discovery sessions § Evaluated using classification and ranking metrics 30
  • 31. Row Placement Model: Performance § Metrics § Achieved high classification metrics for predicting continuation vs discovery § Error types: § False positives è CW occupies top of the page unnecessarily § False negative è Difficult for member to find the CW title § Placing the row § Threshold trades off FP and FN è Hard to tune offline § Tuned the threshold by A/B testing 31
  • 32. Reusing the Title Ranking Model § Use the title-level scores § Calibrate scores to get probability Pt of continuation for each CW title t § Aggregate into an overall probability of continuation § E.g., assuming independence: PCW = 1 - ∏tϵCW (1- Pt) § Pro: Avoids maintaining two separate models § Con: Not as accurate as a dedicated model 32
  • 33. Context Awareness § Title ranks highest on the same time of day and device as last play § Experiment: § Played “Sid the Science Kid” on iPhone § Played “Narcos” on the website è Different ranking on iPhone and Web 33
  • 34. Serving the CW Row in Production § Score cannot be precomputed è Real- or near real-time § Some features are context dependent § Row should refresh each time a member watches a title § Need to push updates to clients to keep the row fresh § Latency bottleneck: Data transfers from the cache to computation backend § Requires careful backend engineering § Fallback strategy: If computation fails, can use recency ranking 34
  • 35. Conclusions and Future Directions 35
  • 36. Conclusions § Important to understand different modes of behavior § Continuation is a key driver of streaming hours § Improving CW recommendations improves member experience § A/B testing showed significant boost in user engagement § Future: § Incorporate the placement of CW row (and others) into the main page construction model § When can we automatically start resuming a title? 36
  • 37. Questions? Upcoming blog post on this topic at: techblog.netflix.com Job openings: jobs.netflix.com 37