SlideShare a Scribd company logo
1 of 18
Download to read offline
Building
Recommender
Systems for Fashion
Nick Landia
Dressipi
@dressipi
Dressipi
• Fashion Recommendations & Style Advice
• Going for 5 years
• 25 employees
• B2B
In this talk
• Fashion Domain Characteristics
• Advice & Recommendation Reasons
Fashion Domain
Items: “Fast Fashion”
• Short lifetime of items: a couple of months on average
• Large retailers release new garments daily
Fashion Domain
Users
• Taste changes over time
• Seasons complicate things
• Trends can drastically change user preferences very quickly!
Fashion Domain
What this means for recommenders
• Item catalogue changes rapidly
• High sparsity
• Use content data
• We need recommenders that can recognise changes in user
preference and respond quickly
Advice & Recommendation Reasons
Why do we need an advice component
• Users are looking for guidance and validation that
something will or will not work for them
• Users might not know what fits them best
• Fashion confidence is a big factor – out of comfort zone
recommendations
Advice & Recommendation Reasons
Challenges
• Can’t blindly trust historical interaction data
• Need additional information on users and garments
• Need expert knowledge
• Need to communicate in a user-understandable way
Advice & Recommendation Reasons
How we do this
• Users fill out questionnaire
• We label garments
• Stylists encode fashion rules
Style Rules
“Your key shapes:
The key to dressing your slim silhouette is to disguise your broad
shoulders. Clothes that add flattering volume to your lower half
help to balance out your proportions.
Look for lower necklines that soften you shoulder line. Avoid fussy
shoulder details that add bulk to this area.”
Advice & Recommendation Reasons
• Recommendation reasons have to make sense to the user
• The point of reasons is not to justify “why are you showing
me this”, it is to answer “why is this item good for me”
• Give useful information to the user
Summary
• Item catalogue and user preferences change over time,
faster than in other domains
• It’s not just about product recommendations, users want
advice
• Make recommendation reasons useful to the user
• User-Item interaction data alone is not enough
Learnings
• Start from domain rather than algorithm
• Human-understandable features are great!
• allow for applications that are useful to the user, i.e. advice and
recommendation reasons
• allow for more involvement of domain experts and more rapid
iteration on algorithm development
Thanks!
• We are hiring!
• Open for collaboration with academia!
• nick@dressipi.com

More Related Content

What's hot

Object detection presentation
Object detection presentationObject detection presentation
Object detection presentationAshwinBicholiya
 
Introduction to MERN Stack
Introduction to MERN StackIntroduction to MERN Stack
Introduction to MERN StackSurya937648
 
Book Recommendation Engine
Book Recommendation EngineBook Recommendation Engine
Book Recommendation EngineShravaniBheema
 
Recommendation Systems
Recommendation SystemsRecommendation Systems
Recommendation SystemsRobin Reni
 
Setting up your development environment
Setting up your development environmentSetting up your development environment
Setting up your development environmentNicole Ryan
 
01 elements of modern networking by nader elmansi
01 elements of modern networking by nader elmansi01 elements of modern networking by nader elmansi
01 elements of modern networking by nader elmansiNader Elmansi
 
RecSysTEL lecture at advanced SIKS course, NL
RecSysTEL lecture at advanced SIKS course, NLRecSysTEL lecture at advanced SIKS course, NL
RecSysTEL lecture at advanced SIKS course, NLHendrik Drachsler
 
IoT Levels and Deployment Templates
IoT Levels and Deployment TemplatesIoT Levels and Deployment Templates
IoT Levels and Deployment TemplatesPrakash Honnur
 
Chapter 5 IoT Design methodologies
Chapter 5 IoT Design methodologiesChapter 5 IoT Design methodologies
Chapter 5 IoT Design methodologiespavan penugonda
 
Recommendation system
Recommendation system Recommendation system
Recommendation system Vikrant Arya
 
Recommender system introduction
Recommender system   introductionRecommender system   introduction
Recommender system introductionLiang Xiang
 
Content based recommendation systems
Content based recommendation systemsContent based recommendation systems
Content based recommendation systemsAravindharamanan S
 
Web crawler with seo analysis
Web crawler with seo analysis Web crawler with seo analysis
Web crawler with seo analysis Vikram Parmar
 
Movie recommendation project
Movie recommendation projectMovie recommendation project
Movie recommendation projectAbhishek Jaisingh
 
The Full Stack Web Development
The Full Stack Web DevelopmentThe Full Stack Web Development
The Full Stack Web DevelopmentSam Dias
 
Deep Learning in Recommender Systems - RecSys Summer School 2017
Deep Learning in Recommender Systems - RecSys Summer School 2017Deep Learning in Recommender Systems - RecSys Summer School 2017
Deep Learning in Recommender Systems - RecSys Summer School 2017Balázs Hidasi
 

What's hot (20)

Object detection presentation
Object detection presentationObject detection presentation
Object detection presentation
 
Introduction to MERN Stack
Introduction to MERN StackIntroduction to MERN Stack
Introduction to MERN Stack
 
Book Recommendation Engine
Book Recommendation EngineBook Recommendation Engine
Book Recommendation Engine
 
Recommendation Systems
Recommendation SystemsRecommendation Systems
Recommendation Systems
 
Setting up your development environment
Setting up your development environmentSetting up your development environment
Setting up your development environment
 
01 elements of modern networking by nader elmansi
01 elements of modern networking by nader elmansi01 elements of modern networking by nader elmansi
01 elements of modern networking by nader elmansi
 
RecSysTEL lecture at advanced SIKS course, NL
RecSysTEL lecture at advanced SIKS course, NLRecSysTEL lecture at advanced SIKS course, NL
RecSysTEL lecture at advanced SIKS course, NL
 
IoT Levels and Deployment Templates
IoT Levels and Deployment TemplatesIoT Levels and Deployment Templates
IoT Levels and Deployment Templates
 
6LoWPAN: An open IoT Networking Protocol
6LoWPAN: An open IoT Networking Protocol6LoWPAN: An open IoT Networking Protocol
6LoWPAN: An open IoT Networking Protocol
 
Chapter 5 IoT Design methodologies
Chapter 5 IoT Design methodologiesChapter 5 IoT Design methodologies
Chapter 5 IoT Design methodologies
 
Recommendation system
Recommendation system Recommendation system
Recommendation system
 
Recommender Systems
Recommender SystemsRecommender Systems
Recommender Systems
 
Webservices
WebservicesWebservices
Webservices
 
Php Presentation
Php PresentationPhp Presentation
Php Presentation
 
Recommender system introduction
Recommender system   introductionRecommender system   introduction
Recommender system introduction
 
Content based recommendation systems
Content based recommendation systemsContent based recommendation systems
Content based recommendation systems
 
Web crawler with seo analysis
Web crawler with seo analysis Web crawler with seo analysis
Web crawler with seo analysis
 
Movie recommendation project
Movie recommendation projectMovie recommendation project
Movie recommendation project
 
The Full Stack Web Development
The Full Stack Web DevelopmentThe Full Stack Web Development
The Full Stack Web Development
 
Deep Learning in Recommender Systems - RecSys Summer School 2017
Deep Learning in Recommender Systems - RecSys Summer School 2017Deep Learning in Recommender Systems - RecSys Summer School 2017
Deep Learning in Recommender Systems - RecSys Summer School 2017
 

Viewers also liked

Recommendations for Building Machine Learning Software
Recommendations for Building Machine Learning SoftwareRecommendations for Building Machine Learning Software
Recommendations for Building Machine Learning SoftwareJustin Basilico
 
Lessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixLessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixJustin Basilico
 
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...Benjamin Le
 
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...Justin Basilico
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectiveJustin Basilico
 
Personalization Challenges in E-Learning
Personalization Challenges in E-LearningPersonalization Challenges in E-Learning
Personalization Challenges in E-LearningRoberto Turrin
 
Personalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing RecommendationsPersonalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing RecommendationsJustin Basilico
 
Bootstrapping a Destination Recommendation Engine
Bootstrapping a Destination Recommendation EngineBootstrapping a Destination Recommendation Engine
Bootstrapping a Destination Recommendation EngineNeal Lathia
 
Déjà Vu: The Importance of Time and Causality in Recommender Systems
Déjà Vu: The Importance of Time and Causality in Recommender SystemsDéjà Vu: The Importance of Time and Causality in Recommender Systems
Déjà Vu: The Importance of Time and Causality in Recommender SystemsJustin Basilico
 

Viewers also liked (9)

Recommendations for Building Machine Learning Software
Recommendations for Building Machine Learning SoftwareRecommendations for Building Machine Learning Software
Recommendations for Building Machine Learning Software
 
Lessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixLessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at Netflix
 
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...
Personalized Job Recommendation System at LinkedIn: Practical Challenges and ...
 
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...
Is that a Time Machine? Some Design Patterns for Real World Machine Learning ...
 
Past, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry PerspectivePast, Present & Future of Recommender Systems: An Industry Perspective
Past, Present & Future of Recommender Systems: An Industry Perspective
 
Personalization Challenges in E-Learning
Personalization Challenges in E-LearningPersonalization Challenges in E-Learning
Personalization Challenges in E-Learning
 
Personalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing RecommendationsPersonalized Page Generation for Browsing Recommendations
Personalized Page Generation for Browsing Recommendations
 
Bootstrapping a Destination Recommendation Engine
Bootstrapping a Destination Recommendation EngineBootstrapping a Destination Recommendation Engine
Bootstrapping a Destination Recommendation Engine
 
Déjà Vu: The Importance of Time and Causality in Recommender Systems
Déjà Vu: The Importance of Time and Causality in Recommender SystemsDéjà Vu: The Importance of Time and Causality in Recommender Systems
Déjà Vu: The Importance of Time and Causality in Recommender Systems
 

Similar to Building Recommender Systems for Fashion

Customer Service and Support
Customer Service and SupportCustomer Service and Support
Customer Service and SupportAnand Shah PMP
 
Dressipi - Personalised recommendation engine for fashion consumers
Dressipi - Personalised recommendation engine for fashion consumersDressipi - Personalised recommendation engine for fashion consumers
Dressipi - Personalised recommendation engine for fashion consumersProject Juno
 
Service frameworks and toolkits: Making design artefacts actionable
Service frameworks and toolkits: Making design artefacts actionableService frameworks and toolkits: Making design artefacts actionable
Service frameworks and toolkits: Making design artefacts actionableKarina Smith
 
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)Carolyn S Dean
 
Building stronger content teams - Melissa Breker at boye 18
Building stronger content teams - Melissa Breker at boye 18Building stronger content teams - Melissa Breker at boye 18
Building stronger content teams - Melissa Breker at boye 18Boye & Co
 
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...Dakiry
 
lean construction and integrated project delivery
lean construction and integrated project delivery lean construction and integrated project delivery
lean construction and integrated project delivery Mahendra Bhuva
 
Customer to Product Idea Iteration by Amazon's Product Manager
Customer to Product Idea Iteration by Amazon's Product ManagerCustomer to Product Idea Iteration by Amazon's Product Manager
Customer to Product Idea Iteration by Amazon's Product ManagerProduct School
 
getting feedback 360 right
getting feedback 360 right getting feedback 360 right
getting feedback 360 right neha singh
 
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...AppFolio
 
User Research When You Can't Reach Your Users 20141016
User Research When You Can't Reach Your Users 20141016User Research When You Can't Reach Your Users 20141016
User Research When You Can't Reach Your Users 20141016Heather Staudt
 
Session III final
Session III finalSession III final
Session III finalJake Jacobs
 
Ada online marketing v1.1
Ada online marketing v1.1Ada online marketing v1.1
Ada online marketing v1.1Carolyn S Dean
 

Similar to Building Recommender Systems for Fashion (20)

Customer Service and Support
Customer Service and SupportCustomer Service and Support
Customer Service and Support
 
NAATP
NAATPNAATP
NAATP
 
Dressipi - Personalised recommendation engine for fashion consumers
Dressipi - Personalised recommendation engine for fashion consumersDressipi - Personalised recommendation engine for fashion consumers
Dressipi - Personalised recommendation engine for fashion consumers
 
The no authority CAD Manager
The no authority CAD ManagerThe no authority CAD Manager
The no authority CAD Manager
 
Service frameworks and toolkits: Making design artefacts actionable
Service frameworks and toolkits: Making design artefacts actionableService frameworks and toolkits: Making design artefacts actionable
Service frameworks and toolkits: Making design artefacts actionable
 
Chapter03
Chapter03Chapter03
Chapter03
 
Requirements elicitation
Requirements elicitationRequirements elicitation
Requirements elicitation
 
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)
Dental Marketing Online - Trends And Best Practice In Online Marketing (ADX16)
 
Building stronger content teams - Melissa Breker at boye 18
Building stronger content teams - Melissa Breker at boye 18Building stronger content teams - Melissa Breker at boye 18
Building stronger content teams - Melissa Breker at boye 18
 
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...
Романа Косцик “New project begins. Jump in and keep calm. Everything will be ...
 
lean construction and integrated project delivery
lean construction and integrated project delivery lean construction and integrated project delivery
lean construction and integrated project delivery
 
Customer to Product Idea Iteration by Amazon's Product Manager
Customer to Product Idea Iteration by Amazon's Product ManagerCustomer to Product Idea Iteration by Amazon's Product Manager
Customer to Product Idea Iteration by Amazon's Product Manager
 
Social media marketing 8 22-13
Social media marketing 8 22-13Social media marketing 8 22-13
Social media marketing 8 22-13
 
HCI_Lecture04.pptx
HCI_Lecture04.pptxHCI_Lecture04.pptx
HCI_Lecture04.pptx
 
getting feedback 360 right
getting feedback 360 right getting feedback 360 right
getting feedback 360 right
 
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...
Hire Smart: Why Your Recruiting Process Can Make or Break Your Business (Prop...
 
User Research When You Can't Reach Your Users 20141016
User Research When You Can't Reach Your Users 20141016User Research When You Can't Reach Your Users 20141016
User Research When You Can't Reach Your Users 20141016
 
Session III final
Session III finalSession III final
Session III final
 
Ada online marketing v1.1
Ada online marketing v1.1Ada online marketing v1.1
Ada online marketing v1.1
 
Marketing the Gold
Marketing the GoldMarketing the Gold
Marketing the Gold
 

Building Recommender Systems for Fashion

  • 2. Dressipi • Fashion Recommendations & Style Advice • Going for 5 years • 25 employees • B2B
  • 3.
  • 4. In this talk • Fashion Domain Characteristics • Advice & Recommendation Reasons
  • 5. Fashion Domain Items: “Fast Fashion” • Short lifetime of items: a couple of months on average • Large retailers release new garments daily
  • 6. Fashion Domain Users • Taste changes over time • Seasons complicate things • Trends can drastically change user preferences very quickly!
  • 7. Fashion Domain What this means for recommenders • Item catalogue changes rapidly • High sparsity • Use content data • We need recommenders that can recognise changes in user preference and respond quickly
  • 8. Advice & Recommendation Reasons Why do we need an advice component • Users are looking for guidance and validation that something will or will not work for them • Users might not know what fits them best • Fashion confidence is a big factor – out of comfort zone recommendations
  • 9. Advice & Recommendation Reasons Challenges • Can’t blindly trust historical interaction data • Need additional information on users and garments • Need expert knowledge • Need to communicate in a user-understandable way
  • 10. Advice & Recommendation Reasons How we do this • Users fill out questionnaire • We label garments • Stylists encode fashion rules
  • 12.
  • 13. “Your key shapes: The key to dressing your slim silhouette is to disguise your broad shoulders. Clothes that add flattering volume to your lower half help to balance out your proportions. Look for lower necklines that soften you shoulder line. Avoid fussy shoulder details that add bulk to this area.”
  • 14.
  • 15. Advice & Recommendation Reasons • Recommendation reasons have to make sense to the user • The point of reasons is not to justify “why are you showing me this”, it is to answer “why is this item good for me” • Give useful information to the user
  • 16. Summary • Item catalogue and user preferences change over time, faster than in other domains • It’s not just about product recommendations, users want advice • Make recommendation reasons useful to the user • User-Item interaction data alone is not enough
  • 17. Learnings • Start from domain rather than algorithm • Human-understandable features are great! • allow for applications that are useful to the user, i.e. advice and recommendation reasons • allow for more involvement of domain experts and more rapid iteration on algorithm development
  • 18. Thanks! • We are hiring! • Open for collaboration with academia! • nick@dressipi.com