Jason Bell

Slides from Jason Bell's recent presentation at 'Bash', a semi-regular developer event held in Belfast (http://instil.co/2013/09/18/shiny-bash/).

Introduction to Hadoop and Machine Learning

Jason is technical architect for SportsFusion, he's previously worked with Learning Pool, AirPOS, The Press Association and others in Northern Ireland, England and California. His main areas of expertise lie in Java, Hadoop and RabbitMQ messaging with a large leaning on customer prediction data and analytics. He also teaches core programming concepts for the University of Ulster and is also working on a book on Machine Learning for a 2014 release.

Transcript

1. Hadoop &! Machine Learning @hadooping
2. Rules of the road....
3. Whatever happens...
4. tweet
5. tweet (or whatever floats yer boat)
6. What is Hadoop?! ! ! !
7. What is Hadoop?! A real quick overview (AP)! ! !
8. What is Hadoop?! A real quick overview (AP)! Word count == Hello World!! !
9. What is Hadoop?! A real quick overview (AP)! Word count == Hello World! ! A more useful example!
10. What is Hadoop?! A real quick overview (AP)! Word count == Hello World! ! A more useful example! Machine Learning!
11. What is Hadoop?! A real quick overview (AP)! Word count == Hello World! ! A more useful example! Machine Learning! Q & A
12. >> What is Hadoop?! ! ! !
13. An open source framework for storing and large scale processing of data sets on clusters of commodity hardware.! !
14. ! >>Is it shiny and hot?! ! !
15. Depends who you ask....
17. Clusters == lots of processing...! !
18. It goes to 11.! !
19. Uses MapReduce wonderfully.! !
20. >>A real quick overview (AP)
21. AP = Audience Participation
22. AP = Audience Participation! ! (and I know you’ll all hate me for it but I really don’t care that much, we’ll still be friends....)
23. AP = Audience Participation! A quick sort!
24. AP = Audience Participation! A quick sort with MapReduce!
25. >>WordCount == Hello World!!
26. Basic MapReduce in Action! WordCount!
27. Basic MapReduce in Action! WordCount! It’s boring but explains it well.
28. Basic MapReduce in Action! WordCount!
29. Map in parallel
30. >>A more useful example
31. >>A more useful example! ! The coffee shop....
32. 20,000 customers with ! 13 month sales data 1,3,11,6,10,7,10,12,9,7,6,10,14,5
33. 20,000 customers 1. The average 12 month sales.! 2. The variance! 3. The month 13 sales drop against the average! 4. Number months which are 40% below avg.
34. 20,000 customers So far we’ve only output to one file. Now I want to segment my customers into groups I can market to.
35. >>Recommending Stuff 1000 Users, 5000 Items ! and ! 100,000 Recommendations
36. >>Recommending Stuff 196 242 3 881250949 186 302 3 891717742 22 377 1 878887116 244 51 2 880606923
37. >>Q&A? ! >>Thank you! >>@hadooping! >>http://github.com/jasebell