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.

Introduction to Recommendation Systems (Vietnam Web Submit)

591 views

Published on

1) Why do we need recommendation systems ?
2) How can we think with recommendation systems ?
3) How can we implement a recommendation system with open source technologies ?
RFX framework https://github.com/rfxlab
Apache Kafka: https://kafka.apache.org
Apache Spark: https://spark.apache.org

Published in: Data & Analytics
  • Hello! Get Your Professional Job-Winning Resume Here - Check our website! https://vk.cc/818RFv
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here

Introduction to Recommendation Systems (Vietnam Web Submit)

  1. 1. Introduction to Recommendation Systems Two Decades of Challenge By Trieu Nguyen - Head of Platform at Blueseed Digital My personal email: tantrieuf31@gmail.com
  2. 2. Agenda 1. Why do we need recommendation systems ? 2. How can we think with recommendation systems ? 3. How can we implement a recommendation system with open source technologies ? a. RFX framework https://github.com/rfxlab b. Apache Kafka: https://kafka.apache.org c. Apache Spark: https://spark.apache.org
  3. 3. 1 - Why do we need recommendation systems
  4. 4. Recommendation Engine — Examples ● Facebook — “People You May Know” ● Netflix — “Other Movies You May Enjoy” ● LinkedIn — “Jobs You May Be Interested In” ● Amazon — “Customer who bought this item also bought …” ● Google — “Visually Similar Images” ● YouTube — “Recommended Videos” ● Tinder — ”Decide who you want to date” ...
  5. 5. “Recommended for you” by YouTube User-based Collaboration Filter Item-based Collaboration Filter
  6. 6. Amazon Recommendation Engine
  7. 7. https://www.computer.org/csdl/mags/ic/2017/03/mic2017030012.html
  8. 8. That my books, mostly from Amazon Recommendations :)
  9. 9. https://www.slideshare.net/xamat/qcon-sf-2013-machine-learning-recommender-systems-netflix-scale At Netflix
  10. 10. #ChoiceArchitecture is the basic idea behind recommendation
  11. 11. Why Should We Use Recommendation Engines? 1. Two-thirds of movies watched by Netflix customers are recommended movies 2. 38% of click-through rates on Google News are recommended links 3. 35% of sales at Amazon arise from recommended products
  12. 12. The value of product recommendation 1. Retain user loyalty 2. Builds the volume of user traffic 3. Delivers more convenient UX to your user 4. Give your business a wider marketplace
  13. 13. So what is recommendation engine ? In technical terms, a recommendation engine problem is to develop a mathematical model or objective function which can predict how much a user will like an item. If U = {users}, I = {items} then F = Objective function and measures the usefulness of item I to user U, given by: F: U x I → R Where R = {recommended items}. For each user u, we want to choose the item i that maximizes the objective function:
  14. 14. 2 - How can we design recommendation systems?
  15. 15. Popular techniques to build recommendation systems 1. Collaborative Filtering 2. Content-Based Filtering 3. Hybrid Recommendation Systems
  16. 16. User-based Collaborative Filtering ● Based on a large amount of information on users’ behaviors, activities or preferences ● Predicting what users will like based on their similarity to other users. ● A key advantage: ○ accurately recommending complex items such as movies without requiring an “understanding” of the item itself
  17. 17. User-based Collaborative Filtering
  18. 18. Content Based Filtering ( Item-based collaborative filtering) Key ideas: ● Based on a description of the item and a profile of the user’s preference. ● Keywords are used to describe the items ● User profile is built to indicate the type of item this user likes.
  19. 19. Content Based Filtering ( Item-based collaborative filtering)
  20. 20. Hybrid Recommendation Systems These methods can be used to overcome some of the common problems: 1. cold start (no information on users’ behaviors) 2. sparsity problem (dense matrix)
  21. 21. Hybrid Recommendation Systems
  22. 22. Example of Hybrid Recommendation Systems User-based Collaboration Filter Item-based Collaboration Filter
  23. 23. User-based VS Item-based
  24. 24. 3 - Building Data Pipeline for Video Recommendation System
  25. 25. Simple version of Video Data Pipeline https://github.com/rfxlab/rfx-video-analytics
  26. 26. Simple Recommendation Engine with Apache Spark MLLib
  27. 27. Picture from http://www.rfxlab.com/p/ad.html
  28. 28. https://www.infoq.com/news/2016/09/How-YouTube-Recommendation-Works
  29. 29. The final question: How computers could know what we really want ?
  30. 30. Follow this page to get more information http://BigDataVietnam.org https://facebook.com/BigDataVN
  31. 31. Ref links about Apache Spark http://blogs.quovantis.com/recommendation-engine-using-apache-spark/ https://chimpler.wordpress.com/2014/07/22/building-a-food-recommendation -engine-with-spark-mllib-and-play http://ampcamp.berkeley.edu/big-data-mini-course/movie-recommendation-wi th-mllib.html https://stanford.edu/~rezab/classes/cme323/S16/ https://www.codementor.io/jadianes/building-a-recommender-with-apache-sp ark-python-example-app-part1-du1083qbw https://bugra.github.io/work/notes/2014-04-19/alternating-least-squares-meth od-for-collaborative-filtering/
  32. 32. Ref links http://dataconomy.com/2015/03/an-introduction-to-recommendation-engines https://www.tastehit.com/blog/personal-data-in-personalization-and-advertising/ https://medium.com/@humansforai/recommendation-engines-e431b6b6b446 http://infolab.stanford.edu/~ullman/mmds/ch9.pdf https://medium.com/netflix-techblog/netflix-recommendations-beyond-the-5-stars- part-1-55838468f429

×