Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Past present and future of Recommender Systems: an Industry Perspective


Published on

Talk presented at the ACM Recommender Systems Conference in Boston, 2016

Published in: Technology
  • Morgan . I can see what your saying... Troy `s postlng is flabbergasting... on thursday I got a gorgeous Ford since I been earnin $6276 this last month an would you believe ten grand this past-munth . it's actualy the coolest job I have ever done . I started this nine months/ago and practically straight away began to make more than $69 per hour . Get More Info ►►►►►►►►►
    Are you sure you want to  Yes  No
    Your message goes here

Past present and future of Recommender Systems: an Industry Perspective

  1. 1. 11 Past, Present & Future of Recommender Systems: An Industry Perspective Xavier Amatriain (Quora) Justin Basilico (Netflix) RecSys 2016 @xamat @JustinBasilico DeLorean image by &
  2. 2. 2 1. Past
  3. 3. 3 Netflix Prize 2006
  4. 4. 4 For more information ...
  5. 5. 5 2. Present
  6. 6. 6 Recommender Systems in Industry Recommender Systems are used pervasively across application domains
  7. 7. 7 Recommender Systems in Industry click upvote downvote expand share
  8. 8. 8 Beyond explicit feedback ▪ Applications typically oriented around an action: click, buy, read, listen, watch, … ▪ Implicit Feedback ▪ More data: Implicit feedback comes as part of normal use ▪ Better data: Matches with actions we want to predict ▪ Augment with contextual information ▪ Content for cold-start ▪ Hybrid: Combine together when you can
  9. 9. 9 Ranking ▪ Ranking items is central to recommending ▪ News feeds ▪ Items in catalogs ▪ … ▪ Most recsys can be assimilated to: ▪ A learning-to-rank approach ▪ A feature engineering problem
  10. 10. 10 Everything is a RecommendationRows Ranking
  11. 11. 11 3. Future
  12. 12. 12 Many interesting future directions 1. Indirect feedback 2. Value-awareness 3. Full-page optimization 4. Personalizing the how ▪ Others ▪ Intent/session awareness ▪ Interactive recommendations ▪ Context awareness ▪ Deep learning for recommendations ▪ Conversational interfaces/bots for recommendations ▪ …
  13. 13. 13 Indirect Feedback Challenges ▪ User can only click on what you show ▪ But, what you show is the result of what your model predicted is good ▪ No counterfactuals ▪ Implicit data has no real “negatives” Potential solutions ▪ Attention models ▪ Context is also indirect/implicit feedback ▪ Explore/exploit approaches and learning across time ▪ ... click upvote downvote expand share
  14. 14. 14 Value-aware recommendations ▪ Recsys optimize for probability of action ▪ Not all clicks/actions have the same “reward” ▪ Different margin in ecommerce ▪ Different “quality” of content ▪ Long-term retention vs. short-term clicks (clickbait) ▪ … ▪ In Quora, the value of showing a story to a user is approximated by weighted sum of actions: v = ∑a va 1{ya = 1} ▪ Extreme application of value-aware recommendations: suggest items to create that have the highest value ▪ Netflix: Which shows to produce or license ▪ Quora: Answers and questions that are not in the service
  15. 15. 15 Full page optimization ▪ Recommendations are rarely displayed in isolation ▪ Rankings are combined with many other elements to make a page ▪ Want to optimize the whole page ▪ Means jointly solving for set of items and their placement ▪ While incorporating ▪ Diversity, freshness, exploration ▪ Depth and coverage of the item set ▪ Non-recommendation elements (navigation, editorial, etc.) ▪ Needs work hand-in-hand with the UX
  16. 16. 16 Personalizing How We Recommend (… not just what we recommend) ▪ Algorithm level: Ideal balance of diversity, novelty, popularity, freshness, etc. may depend on the person ▪ Display level: How you present items or explain recommendations can also be personalized ▪ Select the best information and presentation for a user to quickly decide whether or not they want an item ▪ Interaction level: Balancing the needs of lean-back users and power users
  17. 17. 17 Rows Example: Rows & Beyond Hero Image Predicted rating Evidence Synopsis Horizontal Image Row Title Metadata Ranking
  18. 18. 18 4. Conclusions
  19. 19. 19 Conclusions ▪ Approaches have evolved a lot in the past 10 years ▪ Looking forward to the next 10 ▪ Industry and academia working together has advanced the field since the beginning, we should make sure that continues
  20. 20. 20 Thank You Justin Basilico @JustinBasilico Xavier Amatriain @xamat