[aws la media & entertainment event 2015]: cloud analytics for audience engagement

44
©2015, Amazon Web Services, Inc. or its affiliates. All rights reserved Cloud Analytics for Audience Engagement Mike Limcaco | AWS Principal Solutions Architect Mick Bass | 47Lining CEO

Upload: amazon-web-services

Post on 22-Feb-2017

1.290 views

Category:

Technology


3 download

TRANSCRIPT

Page 1: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

©2015, Amazon Web Services, Inc. or its affiliates. All rights reserved

Cloud Analytics for Audience Engagement

Mike Limcaco | AWS Principal Solutions Architect Mick Bass | 47Lining CEO

Page 2: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

http://mashable.com

Page 3: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Search

Watch

Listen

Play

Download

Purchase

Rate It

Review It

Sharing

Tagging

Bookmarking

Page 4: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

GB TB PB

ZB

EB

Page 5: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

65M+ Subscribers 50 Countries

1000+ Devices 10B+ Hours/Quarter

Page 6: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Netflix: Over 75% of what people watch come from recommendations [1]

[1] http://bit.ly/1WIInoh

Page 7: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Machine Learning for Predictive Analytics

•  Branch of Artificial Intelligence and Statistics •  Programming computers based on historical experience •  Focuses on prediction based on known properties learned

from training data

Signals Predictions

Page 8: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

A Few Machine Learning Tasks

Recommendations

Clustering

Classification

Page 9: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Why Cloud?

•  Scale •  Adaptability •  Agility

Page 10: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

(Some) Big Data Services

Amazon S3

Internet scale storage

Amazon Machine Learning

Hosted predictive analytics service

Amazon EMR

Hosted Hadoop framework

Page 11: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Maximizing Audience Engagement

•  Recommendations

•  Ad / Marketing Campaign Targeting –  Sentiment analysis –  Automated segmentation

•  Predict audience churn

Page 12: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Examples

Page 13: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Recommendations

Page 14: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

The Concept

ü Capture Audience Signal Data ü Create history of user and item preferences

ü Estimate similar users and items ü Record these in Search Engine ü Query Search Engine with User History

ü  Enjoy recommendations!

http://www.slideshare.net/tdunning/recommendation-techn

Page 15: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Log Storage

ETL

User Interface

Serving Layer

users

Recommend Engine

Page 16: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

users

Media platforms

Mobile

Search Play Buy Rate

Recommendations

Page 17: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Apache Mahout on Amazon EMR

•  Library of scalable machine-learning algorithms

•  Single-node as well as distributed capabilities

–  Hadoop Map-Reduce (BATCH | OFFLINE) –  Evolving to support other execution platforms

(NEAR-REALTIME) •  Apache Spark •  H20

Spark H20

Recommendation Clustering Classification

Math Library

Hadoop

Map-Reduce

Page 18: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

mike,view,movie-a mike,view,movie-b mike,view,movie-c mike,buy,movie-b chris,view,movie-b chris,buy,movie-d …

movie-b movie-c:2.772588722239781 movie-a:2.772588722239781

movie-d ….Indicators

(“Items Similar To This….”)

% mahout spark-itemsimilarity -i input-folder/data.txt -o output-folder/ --filter1 buy -fc 1 -ic 2 --filter2 view

Page 19: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 1: Logs à History Matrix

User1 Thing1 User2 Thing2 User3 Thing3 User2 Thing4 User5 Thing1 User1 Thing2 User1 Thing3

Mike

Jon

Mary

Phil

Kris

Logs History Matrix

Page 20: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 2: Estimate Similar Things

History Matrix

2 8

2 4

8

4

Item-Item Matrix

Page 21: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 3: Reduce to Interesting Pairs

2 8

2 4

8

4

Item-Item Matrix

LLR

Indicators (“Items Similar To This….”)

Page 22: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 3: Reduce to Interesting Pairs

Indicators (“Items Similar To This….”)

Items Similar To This

Page 23: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 4: Store Indicators in a Search Engine (BATCH)

Superman Highlander, Dune

Star Wars Raiders, Minority Report

Highlander Superman Mulan Home Alone,

Mermaid Star Trek … … …

4587 223, 5234 748 5345, 235 12 8234 245 9543, 7673 3456 4587 … …

Index

Page 24: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Indicators

Page 25: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Step 5: Query Search Engine w/ User History (REALTIME)

748 Star Wars 45, 235 12 Highlander 8234 245 Mulan 9543,

7673 4587 Superman 12, 5234 3456 Star Trek 2458 …

Query

“12”

5345

3456

12

Page 26: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement
Page 27: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Sentiment Analysis

Page 28: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

“I thought Episode 29 was not without merit J”

Page 29: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Is this Positive or Negative?

Page 30: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

The Concept

ü Capture social media signals ü NRT Streams (Tweets, Comments on FB, IMDB, YouTube …)

ü Push through Sentiment Analyzer ü  Tokenize ü Classify (estimate) as Positive | Negative

ü Provide actionable insight ü  Improve sort order of recommendations ü  Alert / advise Digital Marketing team

Page 31: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Training

Positive Negative

Knowledge Base

Page 32: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Segment Classify

Segment Classify

Segment Classify

Model Training

Positive Negative

Stream Ingest

Stream Ingest

Stream Ingest

Knowledge Base

Page 33: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Segment Classify

Segment Classify

Segment Classify

Model Training

Positive Negative

Stream Ingest

Stream Ingest

Stream Ingest

Knowledge Base

“I adored this movie”

“adore” = POSITIVE

Page 34: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

GNIP

Datasift

Other

Positive Negative

Amazon Kinesis

Amazon Kinesis

Amazon Kinesis

Model Training & Storage

Stream Ingest

Sentiment Classification

Amazon Machine Learning

Trending Sentiment

Neutral Negative Positive

Page 35: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Sentiment Training Sets

Page 36: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Case Study: OTT Predictive Analytics

Page 37: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

UsingDataforStrategicAdvantage

Proof-of-ConceptOTTservicesprovidetheopportunitytoestablisha“conversa.onwiththecustomer.”BigdatatechniquescanbeappliedtofusemassiveamountsofOTTlogs,3rdpartydemographics,andsocialmediasen<menttogaininsightintotheaudienceanddeveloppredic.vecapabili.es.47LiningandAmazonWebServicespartneredtodevelopaPoCforanOTTcustomertouseamachine-drivenapproachtoaccuratelypredictuserbehaviorwithintheirconsumervideoapplica<on.

100MConsumerinterac9ons

MillionsofUsers

3rdPartyDemographics

10KTitles

SocialMedia

Fuse/Visualize

5DataSources*

Per-SegmentRecommenda9ons

Predictors Enablers

71%AccurateChurn

Enhanceaudience

engagementthroughrelevantoffers

ResultsScalability

Automa9on

Agility

Cost

effec9veness

MachineLearning

*OrderofMagnitudeforPoCScope

Page 38: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Speed,Agility,Scalability

WhyCloud?

38

AWSMachineLearning.Delivered71%predic=veaccuracyforuserchurnduringPoC.

Hardware.Cloudeconomicsallowustoonlypayforwhatweuse.

RedshiC.Petabyte-scaledatawarehouse.Effec9velyfeedsMachineLearningorEMR.

Elas=cMapReduce.Unparalleledspeedandscaleforbigdatarecommenda9ons.

Managedservicesthat“justwork”providingspeed,agilityandscale

Pre-CloudChallengesNeedtoprocure

hardwareforpeaks

ProprietarymachinelearningsoTwareonlyDatascien9stscanuse

Needtoprocuredata

warehouse

Always-on,on-premisehadoopclusterswith

highmanagementcosts

Page 39: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

PoCArchitecture

39

OTTApp

S3

RedshiT

MachineLearning

Transforma9ons

PredictorServices

BusinessIntelligenceToolsVisualiza9ons&Dashboards

Perio

dic

3rdPartyDemographicData

Sen9mentAnalysis/SocialData

S3

TitleData

AWSMLAPIMahout

ALSRecommender

Elas9cMapReduce

RAnalysisSandboxes

Lessthan~$1Kininfrastructure

AWS CloudFormation

template stack

Logs

Periodic

Automa<onEnvironmentcanbereplicatedusingNucleator,AnsibleandCloudForma9on

Page 40: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

ThePowerofKnowingyourUsers

40

Per-AccountSignatures

ClusteringDimensions

DistanceHeuris9c

ProfilingDimensions

ClusterAnalysishierarchical|model-based

SegmentAnalysis

1

2

n

segments çCohortsè

100MConsumerinterac9ons

MillionsofUsers

3rdPartyDemographics

10kTitles

SocialMedia

AccountSignatureDefini9ons

6dis<nctuserpersonasemerged

*OrderofMagnitudeforPoCScope

Page 41: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

TurningUserKnowledgeintoEngagement

41

1

2

n

segments

Per-SegmentRecommenda9ons

Predictors

71%AccurateChurn

AccountChurnPredictor.Usingtheiden9fiedSegmentswithAmazonMachineLearning,wedevelopedamodeltopredictwhenanaccountisatriskforchurn.Accuratepredic9onsenablemeaningfulofferstousersbeforethisoccurs.

ViewedContentPredictor.

Usingtheiden9fiedSegments,weusedRedshiC,RandAmazonElas=cMapReduce/Mahouttopredictcontentthatusersmayfinddesirablethatalsoachievesengagementobjec9ves.Suchrecommenda9onsenableproac9vesculp9ngofuserbehavior.

Page 42: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Realizinga“S9ckiness”StrategythroughData

42

AWS’speed,agilityandscalabilityareanaturalfitforOTTpredic9veanaly9cs.

CustomerStrategyAc.onWithcapabili<esdemonstratedandde-risked,customerisintegra<ngpredictors

intotheir2016userexperience.De-riskingcost~$1kininfrastructure

71%accuracyinchurnpredic9on Reten9onoffersin2016

Per-PersonaRecommenda9ons Keepusersgluedin2016

PoCwasonlypossibleintheCloud

Knowyourcustomer

S9ckiness=abilitytosteertheconversa9onwithyourcustomer

Page 43: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Summary

43

Page 44: [AWS LA Media & Entertainment Event 2015]: Cloud Analytics for Audience Engagement

Summary

•  Audience demands increasingly personalized content •  We want to understand and predict these needs and

adapt discovery & delivery of media •  Audience interactions can be analyzed to

–  Surface patterns of common behavior –  Estimate or predict audience demand / churn

•  AWS enables tools & techniques for scalable machine learning