Jaccard Index & Recommendable

Overview of the Jaccard Index and the Recommendable Gem

Transcript

1. Jaccard Index Easy Collaborative Filtering
2. Have you ever... • Allowed users to bookmark or like things? • Allowed users to block or hide things? • Allowed users to hate things? • Implemented those things yourself?
3. What if I told you... You were doing it wrong...
4. Recommendable Adds these convenient methods and adds collaborative filtering.
5. Collaborative Filtering?
6. Collaborative Filtering Collaborative filtering (CF) is a technique used by some recommender systems. Collaborative filtering has two senses, a narrow one and a more general one.[1] In general, collaborative filtering is the process of filtering for information or patterns using techniques involving collaboration among multiple agents, viewpoints, data sources, etc.[1] Applications of collaborative filtering typically involve very large data sets. Collaborative filtering methods have been applied to many different kinds of data including sensing and monitoring data - such as in mineral exploration, environmental sensing over large areas or multiple sensors; financial data - such as financial service institutions that integrate many financial sources; or in electronic commerce and web applications where the focus is on user data, etc. The remainder of this discussion focuses on collaborative filtering for user data, although some of the methods and approaches may apply to the other major applications as well. Source: Wikipedia
7. Collaborative Filtering Making predictions about the interests of a user based on the collective preferences of the entire user base.
8. Easy Mode: Jaccard • Calculating the distance between sample sets. • Works on binary systems (likes, dislikes) • Examples: Recommender Systems (Netflix- like), Biological Classifications, any groups really.
9. Relevant XKCD
10. Jawho? • Paul Jaccard • 19th Swiss Botanist & Plant Physiologist • Awesome time for Taxonomy & Classification • Pioneer in Biogeography (Biodiversity Distribution)
11. How?
12. User Similarity
13. Agreements
14. But I Don’t Agree With You!
15. Likes Minus Dislikes
16. Ruby Example
17. Devise-ish Example
18. 0.2? • Bounds are -1..1 • -1. Your polar opposite • 1. Your long lost twin • Joe and Bob are more similar than dissimilar
19. What about mind reading?
20. Recommendable • By David Celis (New Relic) • Written for goodbre.ws • Easy Easy Gem.
21. So, Requirements. • Recommendable < 2.0, ActiveRecord. • >= 2.0, Most popular ORM’s. • Redis • A worker, such as Sidekiq, Resque, etc.
22. How Use?
23. Links • https://github.com/davidcelis/ recommendable • http://davidcel.is/blog/2012/02/07/ collaborative-filtering-with-likes-and- dislikes/