Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

Big wins with small data. PredictionIO in ecommerce - David Jones

There’s a lot of noise about big data and cutting edge algorithms optimisations. Returning to the basics, this presentation shows you might not need as much data as you think to get real world benefits. Learn about machine learning in ecommerce, PredictionIO and how we used off the shelf, well implemented algorithms to get a 71% increase in revenue with an online wine retailer.

  • Identifiez-vous pour voir les commentaires

Big wins with small data. PredictionIO in ecommerce - David Jones

  1. 1. BigWinswithSmallData: PredictionIOinEcommerce byDavidJones(@d_jones)-TechnicalDirector-ResolveDigital
  2. 2. Comic by @d_jones. Inspired by xkcd.
  3. 3. Machinelearning&Ecommerce better.together.
  4. 4. PredictiveAPIs makeitfeellikeyouhaveyourown teamofMLexperts
  5. 5. Personalised recommendations
  6. 6. Abandonedcartemails
  7. 7. Leadscoring
  8. 8. Emailmarketing
  9. 9. BenefitsofMLinEcommerce Morevisitorswithlongersessionduration Improvedretention Higherconversionrates Largerordersize Bottomline,morerevenue.
  10. 10. machinelearninginecommerce deeplyenhancescustomerexperiences
  11. 11. CustomersexpectMLfeatures Eveniftheydon’tknowwhatMLis
  12. 12. Realworldexample: UnitedCellars
  13. 13. Goal:Increaserevenue Solution:Productrecommendations Implementation:PredictiveAPIs
  14. 14. 16kproductviews 60korders 3kproductratings
  15. 15. 79000rowsSmalldata
  16. 16. Anydatathatlinks customerstoproducts
  17. 17. Optimisedatacollection Initiallyhadnodataforloggedoutusers
  18. 18. PredictionIO Opensourcemachinelearningserver Ecommercetemplates Evaluationmetrics Productionready
  19. 19. WhyPredictionIO? ActiveOpenSourceproject Productivityfocused Built-insupportforApacheSparkMLlib OpinionatedDASEarchitecture: DataSourceandPreparator,Algorithm,Serving,Evaluator
  20. 20. Storefront<->ProcessingQueue<->PredictionIO
  21. 21. PAPIProductionTips Monitorperformance Cachewhererealtimeisnotneeded Deriveddataislowrisk
  22. 22. Bricks&mortarretail inspiresanecommerceexperiencewithML
  23. 23. Withwine tastepreferencesarediverse peopleusuallypreferredorwhite
  24. 24. A/BTESTwithGoogleAnalytics
  25. 25. FilterPredictionsAvoidanythingthatlessensrelevance Don’tshowoutofstockproducts
  26. 26. A/BTestResults 45%longeraveragesession 22%increaseinconversionrate 37%increaseinaverageordersize
  27. 27. 71%MoreRevenue
  28. 28. Wheretonext? Predictivefeaturesinmorechannels Introducenewpredictivefeatures
  29. 29. Conclusion
  30. 30. 1 Smalldatacanbeenough Qualityiskey
  31. 31. 2 PredictiveAPIs areawesome
  32. 32. 3 MachineLearning ishighlyeffectivewith Ecommerce
  33. 33. resolve.digital/papis2015 DavidJones,@d_jones

×