SlideShare ist ein Scribd-Unternehmen logo
1 von 28
www.scout24.com
Building Personalized Data Products
From Idea to Product
AWS Pop-up Loft Berlin | October 15th, 2018 | Stephen Wilson & Sebastian Bolz
Outline
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Personalisation and Relevance at Scout24
• Data Science Toolbox
• Innovation and Experimentation
• Modelling with Personal Analytics Clusters (PAC)
• Production Architecture
Personalisation and Relevance at Scout24
Personalisation & Relevance at Scout24
• strategically important
• drives sessions & revenue
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Personalisation and Relevance at Scout24
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Deliver relevance through personalisation
• Greater relevance → Better experience → Increase in trust and perceived quality
• Anticipate needs → Productive engagement
• Lead identification models to automatically identify target groups.
Architectural Overview of Smart Notifications
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Data Science Toolbox
Data Science Toolbox
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Data Science Toolbox
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Personal Analytics Clusters:
− Fully automated
− Configurable
− Data + Notebooks persisted to S3
− Powerful development environment
• EC2 + Docker:
− Use case specific Docker images
• SageMaker:
− Still in the beginnings for us (experimentation, HackWeek etc)
Innovation and Experimentation
Use case: Lead prediction model
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Business need: we would like to automatically predict whether someone is a lead, based on their behaviour on our platform
• Machine learning problem:
− Scoring model (predict probability)
− Similar to CTR models
− Many users, small number of whom engage
Research
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Look at how other companies do it:
− Logistic regression (Google, Facebook, Bing)
• Read widely: academic papers
• Kaggle competitions in similar problem domains
− Factorisation Machines
• Peer discussions in team
• Technical white paper from Criteo:
− FM implementation in production
Experimentation
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Personal Analytics Clusters:
− Explore, play, experiment
− Seamless sampling of production data from our data lake (S3)
− Try things out quickly, fail fast, make changes, iterate.
• Try things out quickly, fail fast, make changes, iterate
Modelling with PAC
Modelling with PAC: Data Preparation
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• We use One Scout User Database (web tracking)
• Encode business logic that determines a user is a lead
• Select lead users and their corresponding events from S3
• Sample equivalent number of non-lead users for balanced dataset
• Disregard very common events and very infrequent ones
• One Hot Encode
Modelling with PAC: Training
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• We used a TensorFlow implementation of Factorisation Machines
− https://github.com/geffy/tffm
• PAC provides flexibility to try out different approaches:
− Data from 1 year
− 6 months
− 1 month
− 15 days
• Easy to compare performance and monitor training times
Modelling with PAC: Local dev + Run in Cloud
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Production Architecture
Production Architecture: Model Training Pipeline
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
OSUD
(S3)
Data Preparator One Hot Encoder Model Trainer
Latest
Model
(S3)
Production Architecture: Smart Notifications
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
UserIDs
Production Architecture
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
Event Event
Production Architecture
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
Cross device IDS
Production Architecture
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
Is user subscribed
Production Architecture
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
Get user event history
Production Architecture
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
Advisor
Subscriptions
(Redis)
OSUD
(DynamoDB)
Brain
Matching DB
(Aurora)
Latest Model
(S3)
Compute score
Publish score
Push notification if
score > threshold
Results
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• A/B Test:
− A are shown notifications based on a rule-based trigger
− B are shown notifications based on the output of the mortgage lead model
• Group B outperformed the rule-based trigger:
− We issued 90% fewer notifications than the trigger group
− 27% better click-rate than trigger group
− 21% less rejection rate than trigger group
− 7 x more leads from the same amount of notifications
• Outcome: ML-based scoring models deliver on relevance and usefulness.
Fitting it all together
Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
• Seamless delivery of services and content
• Relevant and useful
• Reuse
Contact
ImmobilienScout24 GmbH
Andreasstraße 10
10243 Berlin
Stephen Wilson
Fon +49 30 243 01-1686
stephen.wilson@scout24.com
www.scout24.com
Thank you for your attention!
Sebastian Bolz
Fon +49 30 24301-1228
sebastian.bolz@scout24.com
www.scout24.com

Weitere ähnliche Inhalte

Ähnlich wie Building Personalized Data Products - From Idea to Product

Creating a Data Driven Culture with Amazon QuickSight - Technical 201
Creating a Data Driven Culture with Amazon QuickSight - Technical 201Creating a Data Driven Culture with Amazon QuickSight - Technical 201
Creating a Data Driven Culture with Amazon QuickSight - Technical 201Amazon Web Services
 
Architecture of Big Data Solutions
Architecture of Big Data SolutionsArchitecture of Big Data Solutions
Architecture of Big Data SolutionsGuido Schmutz
 
Getting Started with Amazon QuickSight
Getting Started with Amazon QuickSightGetting Started with Amazon QuickSight
Getting Started with Amazon QuickSightAmazon Web Services
 
Modern Business Intelligence and Advanced Analytics
Modern Business Intelligence and Advanced AnalyticsModern Business Intelligence and Advanced Analytics
Modern Business Intelligence and Advanced AnalyticsCollective Intelligence Inc.
 
Building the Artificially Intelligent Enterprise
Building the Artificially Intelligent EnterpriseBuilding the Artificially Intelligent Enterprise
Building the Artificially Intelligent EnterpriseDatabricks
 
Power BI for Big Data and the New Look of Big Data Solutions
Power BI for Big Data and the New Look of Big Data SolutionsPower BI for Big Data and the New Look of Big Data Solutions
Power BI for Big Data and the New Look of Big Data SolutionsJames Serra
 
SPS Vancouver 2018 - What is CDM and CDS
SPS Vancouver 2018 - What is CDM and CDSSPS Vancouver 2018 - What is CDM and CDS
SPS Vancouver 2018 - What is CDM and CDSNicolas Georgeault
 
Customer Insights: It's time to get personal
Customer Insights: It's time to get personalCustomer Insights: It's time to get personal
Customer Insights: It's time to get personalEmpired
 
Customer Insights: It's time to get personal
Customer Insights: It's time to get personalCustomer Insights: It's time to get personal
Customer Insights: It's time to get personalIntergen
 
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K..."Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...Provectus
 
Getting Started with Amazon QuickSight
Getting Started with Amazon QuickSightGetting Started with Amazon QuickSight
Getting Started with Amazon QuickSightAmazon Web Services
 
Microsoft Fabric Introduction
Microsoft Fabric IntroductionMicrosoft Fabric Introduction
Microsoft Fabric IntroductionJames Serra
 
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...Inside Analysis
 
Modern Data Architectures for Business Insights at Scale
Modern Data Architectures for Business Insights at Scale Modern Data Architectures for Business Insights at Scale
Modern Data Architectures for Business Insights at Scale Amazon Web Services
 
Data Discovery and BI - Is there Really a Difference?
Data Discovery and BI - Is there Really a Difference?Data Discovery and BI - Is there Really a Difference?
Data Discovery and BI - Is there Really a Difference?Inside Analysis
 
Smart Process Automation - Connected Brains 2018
Smart Process Automation - Connected Brains 2018Smart Process Automation - Connected Brains 2018
Smart Process Automation - Connected Brains 2018LoQutus
 
Big Data Expo 2015 - Barnsten Why Data Modelling is Essential
Big Data Expo 2015 - Barnsten Why Data Modelling is EssentialBig Data Expo 2015 - Barnsten Why Data Modelling is Essential
Big Data Expo 2015 - Barnsten Why Data Modelling is EssentialBigDataExpo
 
Formulating Power BI Enterprise Strategy
Formulating Power BI Enterprise StrategyFormulating Power BI Enterprise Strategy
Formulating Power BI Enterprise StrategyTeo Lachev
 
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...Databricks
 

Ähnlich wie Building Personalized Data Products - From Idea to Product (20)

Creating a Data Driven Culture with Amazon QuickSight - Technical 201
Creating a Data Driven Culture with Amazon QuickSight - Technical 201Creating a Data Driven Culture with Amazon QuickSight - Technical 201
Creating a Data Driven Culture with Amazon QuickSight - Technical 201
 
Architecture of Big Data Solutions
Architecture of Big Data SolutionsArchitecture of Big Data Solutions
Architecture of Big Data Solutions
 
Meetup Data-science OVH
Meetup Data-science OVHMeetup Data-science OVH
Meetup Data-science OVH
 
Getting Started with Amazon QuickSight
Getting Started with Amazon QuickSightGetting Started with Amazon QuickSight
Getting Started with Amazon QuickSight
 
Modern Business Intelligence and Advanced Analytics
Modern Business Intelligence and Advanced AnalyticsModern Business Intelligence and Advanced Analytics
Modern Business Intelligence and Advanced Analytics
 
Building the Artificially Intelligent Enterprise
Building the Artificially Intelligent EnterpriseBuilding the Artificially Intelligent Enterprise
Building the Artificially Intelligent Enterprise
 
Power BI for Big Data and the New Look of Big Data Solutions
Power BI for Big Data and the New Look of Big Data SolutionsPower BI for Big Data and the New Look of Big Data Solutions
Power BI for Big Data and the New Look of Big Data Solutions
 
SPS Vancouver 2018 - What is CDM and CDS
SPS Vancouver 2018 - What is CDM and CDSSPS Vancouver 2018 - What is CDM and CDS
SPS Vancouver 2018 - What is CDM and CDS
 
Customer Insights: It's time to get personal
Customer Insights: It's time to get personalCustomer Insights: It's time to get personal
Customer Insights: It's time to get personal
 
Customer Insights: It's time to get personal
Customer Insights: It's time to get personalCustomer Insights: It's time to get personal
Customer Insights: It's time to get personal
 
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K..."Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...
"Building a Modern Data platform in the Cloud", Alex Casalboni, AWS Dev Day K...
 
Getting Started with Amazon QuickSight
Getting Started with Amazon QuickSightGetting Started with Amazon QuickSight
Getting Started with Amazon QuickSight
 
Microsoft Fabric Introduction
Microsoft Fabric IntroductionMicrosoft Fabric Introduction
Microsoft Fabric Introduction
 
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...
At the Tipping Point: Considerations for Cloud BI in a Multi-platform BI Ente...
 
Modern Data Architectures for Business Insights at Scale
Modern Data Architectures for Business Insights at Scale Modern Data Architectures for Business Insights at Scale
Modern Data Architectures for Business Insights at Scale
 
Data Discovery and BI - Is there Really a Difference?
Data Discovery and BI - Is there Really a Difference?Data Discovery and BI - Is there Really a Difference?
Data Discovery and BI - Is there Really a Difference?
 
Smart Process Automation - Connected Brains 2018
Smart Process Automation - Connected Brains 2018Smart Process Automation - Connected Brains 2018
Smart Process Automation - Connected Brains 2018
 
Big Data Expo 2015 - Barnsten Why Data Modelling is Essential
Big Data Expo 2015 - Barnsten Why Data Modelling is EssentialBig Data Expo 2015 - Barnsten Why Data Modelling is Essential
Big Data Expo 2015 - Barnsten Why Data Modelling is Essential
 
Formulating Power BI Enterprise Strategy
Formulating Power BI Enterprise StrategyFormulating Power BI Enterprise Strategy
Formulating Power BI Enterprise Strategy
 
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...
Revolutionizing the Legal Industry with Spark, NLP and Azure Databricks at Cl...
 

Mehr von AWS Germany

Analytics Web Day | From Theory to Practice: Big Data Stories from the Field
Analytics Web Day | From Theory to Practice: Big Data Stories from the FieldAnalytics Web Day | From Theory to Practice: Big Data Stories from the Field
Analytics Web Day | From Theory to Practice: Big Data Stories from the FieldAWS Germany
 
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...AWS Germany
 
Modern Applications Web Day | Impress Your Friends with Your First Serverless...
Modern Applications Web Day | Impress Your Friends with Your First Serverless...Modern Applications Web Day | Impress Your Friends with Your First Serverless...
Modern Applications Web Day | Impress Your Friends with Your First Serverless...AWS Germany
 
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...Modern Applications Web Day | Manage Your Infrastructure and Configuration on...
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...AWS Germany
 
Modern Applications Web Day | Container Workloads on AWS
Modern Applications Web Day | Container Workloads on AWSModern Applications Web Day | Container Workloads on AWS
Modern Applications Web Day | Container Workloads on AWSAWS Germany
 
Modern Applications Web Day | Continuous Delivery to Amazon EKS with Spinnaker
Modern Applications Web Day | Continuous Delivery to Amazon EKS with SpinnakerModern Applications Web Day | Continuous Delivery to Amazon EKS with Spinnaker
Modern Applications Web Day | Continuous Delivery to Amazon EKS with SpinnakerAWS Germany
 
Building Smart Home skills for Alexa
Building Smart Home skills for AlexaBuilding Smart Home skills for Alexa
Building Smart Home skills for AlexaAWS Germany
 
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructure
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructureHotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructure
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructureAWS Germany
 
Wild Rydes with Big Data/Kinesis focus: AWS Serverless Workshop
Wild Rydes with Big Data/Kinesis focus: AWS Serverless WorkshopWild Rydes with Big Data/Kinesis focus: AWS Serverless Workshop
Wild Rydes with Big Data/Kinesis focus: AWS Serverless WorkshopAWS Germany
 
Log Analytics with AWS
Log Analytics with AWSLog Analytics with AWS
Log Analytics with AWSAWS Germany
 
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS AWS Germany
 
AWS Programme für Nonprofits
AWS Programme für NonprofitsAWS Programme für Nonprofits
AWS Programme für NonprofitsAWS Germany
 
Microservices and Data Design
Microservices and Data DesignMicroservices and Data Design
Microservices and Data DesignAWS Germany
 
Serverless vs. Developers – the real crash
Serverless vs. Developers – the real crashServerless vs. Developers – the real crash
Serverless vs. Developers – the real crashAWS Germany
 
Query your data in S3 with SQL and optimize for cost and performance
Query your data in S3 with SQL and optimize for cost and performanceQuery your data in S3 with SQL and optimize for cost and performance
Query your data in S3 with SQL and optimize for cost and performanceAWS Germany
 
Secret Management with Hashicorp’s Vault
Secret Management with Hashicorp’s VaultSecret Management with Hashicorp’s Vault
Secret Management with Hashicorp’s VaultAWS Germany
 
Scale to Infinity with ECS
Scale to Infinity with ECSScale to Infinity with ECS
Scale to Infinity with ECSAWS Germany
 
Containers on AWS - State of the Union
Containers on AWS - State of the UnionContainers on AWS - State of the Union
Containers on AWS - State of the UnionAWS Germany
 
Deploying and Scaling Your First Cloud Application with Amazon Lightsail
Deploying and Scaling Your First Cloud Application with Amazon LightsailDeploying and Scaling Your First Cloud Application with Amazon Lightsail
Deploying and Scaling Your First Cloud Application with Amazon LightsailAWS Germany
 

Mehr von AWS Germany (20)

Analytics Web Day | From Theory to Practice: Big Data Stories from the Field
Analytics Web Day | From Theory to Practice: Big Data Stories from the FieldAnalytics Web Day | From Theory to Practice: Big Data Stories from the Field
Analytics Web Day | From Theory to Practice: Big Data Stories from the Field
 
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...
Analytics Web Day | Query your Data in S3 with SQL and optimize for Cost and ...
 
Modern Applications Web Day | Impress Your Friends with Your First Serverless...
Modern Applications Web Day | Impress Your Friends with Your First Serverless...Modern Applications Web Day | Impress Your Friends with Your First Serverless...
Modern Applications Web Day | Impress Your Friends with Your First Serverless...
 
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...Modern Applications Web Day | Manage Your Infrastructure and Configuration on...
Modern Applications Web Day | Manage Your Infrastructure and Configuration on...
 
Modern Applications Web Day | Container Workloads on AWS
Modern Applications Web Day | Container Workloads on AWSModern Applications Web Day | Container Workloads on AWS
Modern Applications Web Day | Container Workloads on AWS
 
Modern Applications Web Day | Continuous Delivery to Amazon EKS with Spinnaker
Modern Applications Web Day | Continuous Delivery to Amazon EKS with SpinnakerModern Applications Web Day | Continuous Delivery to Amazon EKS with Spinnaker
Modern Applications Web Day | Continuous Delivery to Amazon EKS with Spinnaker
 
Building Smart Home skills for Alexa
Building Smart Home skills for AlexaBuilding Smart Home skills for Alexa
Building Smart Home skills for Alexa
 
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructure
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructureHotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructure
Hotel or Taxi? "Sorting hat" for travel expenses with AWS ML infrastructure
 
Wild Rydes with Big Data/Kinesis focus: AWS Serverless Workshop
Wild Rydes with Big Data/Kinesis focus: AWS Serverless WorkshopWild Rydes with Big Data/Kinesis focus: AWS Serverless Workshop
Wild Rydes with Big Data/Kinesis focus: AWS Serverless Workshop
 
Log Analytics with AWS
Log Analytics with AWSLog Analytics with AWS
Log Analytics with AWS
 
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS
Deep Dive into Concepts and Tools for Analyzing Streaming Data on AWS
 
AWS Programme für Nonprofits
AWS Programme für NonprofitsAWS Programme für Nonprofits
AWS Programme für Nonprofits
 
Microservices and Data Design
Microservices and Data DesignMicroservices and Data Design
Microservices and Data Design
 
Serverless vs. Developers – the real crash
Serverless vs. Developers – the real crashServerless vs. Developers – the real crash
Serverless vs. Developers – the real crash
 
Query your data in S3 with SQL and optimize for cost and performance
Query your data in S3 with SQL and optimize for cost and performanceQuery your data in S3 with SQL and optimize for cost and performance
Query your data in S3 with SQL and optimize for cost and performance
 
Secret Management with Hashicorp’s Vault
Secret Management with Hashicorp’s VaultSecret Management with Hashicorp’s Vault
Secret Management with Hashicorp’s Vault
 
EKS Workshop
 EKS Workshop EKS Workshop
EKS Workshop
 
Scale to Infinity with ECS
Scale to Infinity with ECSScale to Infinity with ECS
Scale to Infinity with ECS
 
Containers on AWS - State of the Union
Containers on AWS - State of the UnionContainers on AWS - State of the Union
Containers on AWS - State of the Union
 
Deploying and Scaling Your First Cloud Application with Amazon Lightsail
Deploying and Scaling Your First Cloud Application with Amazon LightsailDeploying and Scaling Your First Cloud Application with Amazon Lightsail
Deploying and Scaling Your First Cloud Application with Amazon Lightsail
 

Kürzlich hochgeladen

IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxMalak Abu Hammad
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 

Kürzlich hochgeladen (20)

IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 

Building Personalized Data Products - From Idea to Product

  • 1. www.scout24.com Building Personalized Data Products From Idea to Product AWS Pop-up Loft Berlin | October 15th, 2018 | Stephen Wilson & Sebastian Bolz
  • 2. Outline Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Personalisation and Relevance at Scout24 • Data Science Toolbox • Innovation and Experimentation • Modelling with Personal Analytics Clusters (PAC) • Production Architecture
  • 4. Personalisation & Relevance at Scout24 • strategically important • drives sessions & revenue Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
  • 5. Personalisation and Relevance at Scout24 Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Deliver relevance through personalisation • Greater relevance → Better experience → Increase in trust and perceived quality • Anticipate needs → Productive engagement • Lead identification models to automatically identify target groups.
  • 6. Architectural Overview of Smart Notifications Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
  • 8. Data Science Toolbox Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
  • 9. Data Science Toolbox Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Personal Analytics Clusters: − Fully automated − Configurable − Data + Notebooks persisted to S3 − Powerful development environment • EC2 + Docker: − Use case specific Docker images • SageMaker: − Still in the beginnings for us (experimentation, HackWeek etc)
  • 11. Use case: Lead prediction model Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Business need: we would like to automatically predict whether someone is a lead, based on their behaviour on our platform • Machine learning problem: − Scoring model (predict probability) − Similar to CTR models − Many users, small number of whom engage
  • 12. Research Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Look at how other companies do it: − Logistic regression (Google, Facebook, Bing) • Read widely: academic papers • Kaggle competitions in similar problem domains − Factorisation Machines • Peer discussions in team • Technical white paper from Criteo: − FM implementation in production
  • 13. Experimentation Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Personal Analytics Clusters: − Explore, play, experiment − Seamless sampling of production data from our data lake (S3) − Try things out quickly, fail fast, make changes, iterate. • Try things out quickly, fail fast, make changes, iterate
  • 15. Modelling with PAC: Data Preparation Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • We use One Scout User Database (web tracking) • Encode business logic that determines a user is a lead • Select lead users and their corresponding events from S3 • Sample equivalent number of non-lead users for balanced dataset • Disregard very common events and very infrequent ones • One Hot Encode
  • 16. Modelling with PAC: Training Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • We used a TensorFlow implementation of Factorisation Machines − https://github.com/geffy/tffm • PAC provides flexibility to try out different approaches: − Data from 1 year − 6 months − 1 month − 15 days • Easy to compare performance and monitor training times
  • 17. Modelling with PAC: Local dev + Run in Cloud Building Personalized Data Products | Stephen Wilson & Sebastian Bolz
  • 19. Production Architecture: Model Training Pipeline Building Personalized Data Products | Stephen Wilson & Sebastian Bolz OSUD (S3) Data Preparator One Hot Encoder Model Trainer Latest Model (S3)
  • 20. Production Architecture: Smart Notifications Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) UserIDs
  • 21. Production Architecture Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) Event Event
  • 22. Production Architecture Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) Cross device IDS
  • 23. Production Architecture Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) Is user subscribed
  • 24. Production Architecture Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) Get user event history
  • 25. Production Architecture Building Personalized Data Products | Stephen Wilson & Sebastian Bolz Advisor Subscriptions (Redis) OSUD (DynamoDB) Brain Matching DB (Aurora) Latest Model (S3) Compute score Publish score Push notification if score > threshold
  • 26. Results Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • A/B Test: − A are shown notifications based on a rule-based trigger − B are shown notifications based on the output of the mortgage lead model • Group B outperformed the rule-based trigger: − We issued 90% fewer notifications than the trigger group − 27% better click-rate than trigger group − 21% less rejection rate than trigger group − 7 x more leads from the same amount of notifications • Outcome: ML-based scoring models deliver on relevance and usefulness.
  • 27. Fitting it all together Building Personalized Data Products | Stephen Wilson & Sebastian Bolz • Seamless delivery of services and content • Relevant and useful • Reuse
  • 28. Contact ImmobilienScout24 GmbH Andreasstraße 10 10243 Berlin Stephen Wilson Fon +49 30 243 01-1686 stephen.wilson@scout24.com www.scout24.com Thank you for your attention! Sebastian Bolz Fon +49 30 24301-1228 sebastian.bolz@scout24.com www.scout24.com